Recent advances in signal amplification strategy based on oligonucleotide and nanomaterials for microRNA detection-a review

Biosensors and Bioelectronics 99 (2018) 612–624

Contents lists available at ScienceDirect

Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios

Recent advances in signal amplification strategy based on oligonucleotide and nanomaterials for microRNA detection-a review

MARK

⁎

Ying-Xu Chena,b, Ke-Jing Huanga,b, , Ke-Xin Niua,b a b

College of Chemistry and Chemical Engineering, Xinyang Normal University, Xinyang 464000, China Henan Province Key Laboratory of Utilization of Non-metallic Mineral in the South of Henan, Xinyang Normal University, Xinyang 464000, China

A R T I C L E I N F O

A BS T RAC T

Keywords: MicroRNAs Signal-amplification strategies Oligonucleotide Nanomaterials Review

MicroRNAs (MiRNAs) play multiple crucial regulating roles in cell which can regulate one third of proteincoding genes. MiRNAs participate in the developmental and physiological processes of human body, while their aberrant adjustment will be more likely to trigger diseases such as cancers, kidney disease, central nervous system diseases, cardiovascular diseases, diabetes, viral infections and so on. What's worse, for the detection of miRNAs, their small size, high sequence similarity, low abundance and difficult extraction from cells impose great challenges in the analysis. Hence, it's necessary to fabricate accurate and sensitive biosensing platform for miRNAs detection. Up to now, researchers have developed many signal-amplification strategies for miRNAs detection, including hybridization chain reaction, nuclease amplification, rolling circle amplification, catalyzed hairpin assembly amplification and nanomaterials based amplification. These methods are typical, feasible and frequently used. In this review, we retrospect recent advances in signal amplification strategies for detecting miRNAs and point out the pros and cons of them. Furthermore, further prospects and promising developments of the signal-amplification strategies for detecting miRNAs are proposed.

1. Introduction Mature microRNAs (miRNAs), a sort of endogenous noncoding RNAs consisted of approximately 19–23 nucleotides (nt), play an essential role in every biological process such as cell proliferation, differentiation, apoptosis and other physiological or pathological processes (Richard et al., 2016). MiRNA was first discovered in Elegans by Lee and his colleagues in 1993 (Wightman et al., 1993). However, they have been kept out of the spotlight for decades due to its' small size and lack of poly-A tails. At present, nearly 2800 human miRNAs are expounded in public repositories. Nevertheless, Londin et al. (2015) revealed the existence of 3707 new human miRNAs expressed throughout the genome which may play pivotal roles in disease etiology. These discoveries suggest that the repertoire of human miRNAs is larger and more varying than may be suggested by the publicly available repositories. The production of mature miRNA encompasses a great deal of RNA processing steps according to the biogenesis of miRNA. The basic process and expression of mature miRNA is outlined in Fig. 1. Firstly, miRNA coding genes are generally transcribed by RNA polymerase II within the nucleus producing large capped and polyadenylated pri-miRNA transcripts. These pri-miRNA transcripts are processed by the RNase III enzyme and Drosha-DGCR8

⁎

to generate an imperfect stem-loop hairpin structure precursor miRNA (pre-miRNA). Those pre-miRNAs have approximately 70–90 nucleotides which are transported from the nucleus into the cytoplasm by the exportin-5. After Dicer processing, the pre-miRNA is transformed into a transient 22 nt mature double stranded (ds) miRNA: miRNA*. The Dicer also processed the unwinding of these miRNA duplexes to the mature sequence. One strand of the functional mature miRNA then associates with other proteins and enzymes to form the miRNAinduced silencing complex (RISC). The RISC complex functions perfectly or imperfectly match with its complementary target messenger RNA (mRNA), and induces target mRNA degradation, translational inhibition or mRNA cleave (João et al., 2015). Hence miRNA acts as regulators of gene expression via this mechanism (Alan et al., 2014). MiRNA imperfects complementary with 3′-untranslated regions of mRNA to produce RNA interference pathway, in which mRNA transcripts can be cleaved by RISC. It goes the further step to interfere the translation of the mRNA and impair the production of protein (Asahiro et al., 2015). The mechanism eventually results in reducing protein level and deep-seated influences on cellular homeostasis. Therefore, miRNA analysis seems important exceedingly. To date, extensive methods for detecting miRNAs have been developed such as northern blot analysis (Válóczi et al., 2004),

Corresponding author at: College of Chemistry and Chemical Engineering, Xinyang Normal University, Xinyang 464000, China. E-mail address: [email protected] (K.-J. Huang).

http://dx.doi.org/10.1016/j.bios.2017.08.036 Received 12 June 2017; Received in revised form 13 August 2017; Accepted 14 August 2017 Available online 18 August 2017 0956-5663/ © 2017 Elsevier B.V. All rights reserved.

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 1. Schematic illustration of miRNA biogenesis and functions.

cells and high sequence similarity among family members (Zhao et al., 2015). Without the elaborate separation and enrichment processes, it is extremely challenged to detect low abundance of miRNAs in real samples (Shen et al., 2015). In an attempt to surmount these difficulties, a number of signal amplification strategies such as hybridization chain reaction (HCR) (Hong et al., 2013), nuclease amplification (Miao et al., 2015a), rolling circle amplification (RCA) (Jiang et al., 2016), catalyzed hairpin assembly (CHA) (Cai et al., 2017) and nanomaterials based techniques (Shuai et al., 2017a) have been developed. Moreover, the integration of two or more signal amplification strategies has been used for the lower detection limit and higher sensitivity (Wei et al., 2016). This article provides an overview on the recent development of research in the field of signal-amplification techniques for miRNA detection. It starts by introducing the basic mechanism of five classical amplification methods and seriatim discuses their advantages and disadvantages. Some basic and novel methods are presented. Finally, the possible challenges and potential opportunities of signal-amplification techniques for miRNA detection are proposed.

polymerase chain reaction (PCR) (Zhang et al., 2013), microarray assay (Clancy et al., 2017), fluorescence (Chi et al., 2017), electrochemiluminescence (ECL) (Peng et al., 2017), photoelectrochemical (Li et al., 2015), surface plasmon resonance (SPR) (Liu et al., 2017a), capillary electrophoresis (Khan et al., 2011), colorimetric measurement (Li et al., 2016b) and electrochemical biosensor (Shuai et al., 2016a). Among them, northern blotting, microarrays and quantitative reverse transcription polymerase chain reaction (qRT-PCR) are the traditional methods for miRNA detection. Many reports have been conducted to evaluate the advantages and disadvantages of these methods. The northern blotting method is considered as the “gold standard” for miRNA, but it encounters the issues of time and sample consuming with low sensitivity and throughput, therefore it is still not very suitable for practical clinic tests. Although microarray technology make multiple miRNA analysis feasible due to high through put screening ability, the requirements of complicated probe, sophisticated instrumentation and trained person limits the application in point-of-care settings. QRT-PCR offers high sensitivity and covers broad dynamic range for miRNA expression profiling. However, the susceptible to contamination and only performance in centralized laboratories limit its application. To overcome these shortcomings, great efforts have been made to develop new techniques by using different signal readout assays including fluorescence, colorimetry, SPR and electrochemistry. Among them, colorimetric assays, without the aid of advanced instrumentation, offer a cost-effective, rapid and convenient option for miRNA detection. SPR, as a biorecognition transducer, transforms the concentration of biomolecule into optical signal attracts researcher's interest owing to its label-free, real-time, and in-situ detection of nucleic acids, proteins, and small analytes (Yang et al., 2016). However, SPR sensors suffer from nonspecific adsorption on the surface of chips. This drawback limits its practical application (Homola et al., 2008). The application of electrochemical sensors for miRNA detection exhibits many advantages such as high sensitivity and selectivity, reliable reproducibility, simple use for continuous onsite analysis, minimal sample preparation, relatively low cost and short time of response (Wu et al., 2014b). However, miRNA is traced in cells and having its own intrinsic characteristics, such as short sequence, vulnerable degradability, low abundance in total RNA samples, relatively low expression levels in

2. The development and applications of miRNA The major clinical challenge of some malignant diseases is that it conceals onset and the difficult to diagnose in early stage diagnostics (Leni et al., 2014). A biomarker can be generally defined as a measurable indicator of a particular disease state or some other physiological state of an organism (Yang et al., 2014). Therefore, specific and sensitive non-invasive biomarkers for the detection of human multifarious serious illnesses are urgently required to reduce the worldwide morbidity and mortality. After years of effort, scientists find that the expression of miRNA can be changed in the blood because of cellular damage and tissue injury, such as in acute myocardial infarction (Gupta et al., 2016), osteoarthritis (Kung et al., 2017), skin fibrosis (Harmanci et al., 2017) and atherosclerosis (Baldán et al., 2016). Unlike intracellular miRNAs, circulating miRNAs release from cells based on their targeted functions. They may shuttle in and out of circulation (Singh et al., 2016). Their discovery has ushered in new approaches for clinical fields and led to a quest for targeted biomarkers. Facts prove that the expression profiles of these circulating miRNAs in 613

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 2. (A) Illustration of the HCR coupling with 3D DNA structure for miRNA detection. (Reproduced with permission from Miao, et al. (2015a). Copyright 2015, The Royal Society of Chemistry); (B) Schematic illustration of electrochemical miRNA biosensor based on CHA and HCR. (Reproduced with permission from Wu, et al. (2014b). Copyright 2014, Elsevier B.V.); (C) Schematic illustration of fluorescent detection of miRNA-141 based on HCR and CHA. (Reproduced with permission from Wei, et al. (2016). Copyright 2015, Elsevier B.V.); (D) Schematic illustration of the electrochemical biosensor for miRNAs based on HCR and hemin. (Reproduced with permission from Xiang, et al. (2014). Copyright 2014, Elsevier B.V.).

serum, plasma, and other body fluids herald immense potential for their use as novel minimally invasive biomarkers in diagnosing and monitoring human diseases because they are tightly correlated with various serious diseases including cancers (Mariàngels and María, 2011; Zen et al., 2012), kidney disease (Li et al., 2010), central nervous system diseases (Jin et al., 2013), cardiovascular diseases (Samanta et al., 2016), diabetes (Zhu et al., 2015), viral infections (FuentesMattei et al., 2017) and so on. Take cancer as a typical example, cancer is one of the greatest causes of human death in the world with more than 10 million new cases every year. In fact, around one in three people will be diagnosed with cancer throughout their lifetime (Roya et al., 2015; Mahin et al., 2015). Researches show that miRNAs act as tumor suppressors through experimental fact that the expression of miRNA is generally down-regulated in tumor cells compared with normal cells. Interestingly, more than half of all genes that encode miRNA are located at fragile sites or in cancer-associated regions of the genome, suggesting that miRNA may have an intimate relationship with cancer (Rajbir et al., 2016; Liu et al., 2015b). Among cancers, colorectal cancer (CRC) is the second most common cancer in the world and hepatocellular carcinoma (HCC) is the third leading cause of cancer deaths worldwide. HCC is a highly aggressive tumor accounting for the most liver cancers. Several studies have begun to examine miRNA deregulation in HCC and hepatitis-related liver diseases. Indeed, the detection and diagnosis of CRC based on a molecular feature implies that a specific expression profile is found in tumor compared to non-tumor normal cells (Slattery et al., 2011; Ayaz et al., 2013). Numerous studies proposed such differences. There are up to 35

miRNAs have been found up or down regulated in CRC cells as compared to non tumor normal cells. Some of miRNAs, such as miRNA-21 and miRNA-20a, are up-regulated in CRC. But some of them including miRNA-191 and miRNA-192 are down regulated. In the same way, several studies have recently reported a relationship between miRNA and HCC. Compared to normal tissues, among miRNA implicated in HCC such as miRNA-21 and miRNA-222, the aberrant levels of miRNA expressions are up regulated (Wong et al., 2008; Varnholt et al., 2008). However, other miRNAs including miRNA-122a and miRNA-145 decrease in HCC (Gramantieri et al., 2007). Surprisingly, Huang et al. have shown that, among aberrant miRNA in HCC, miRNA-338 affects several clinical features, such as tumor size, tumor–node–metastasis classification, vascular invasion and intrahepatic metastasis (Huang et al., 2009). MiRNA-21 is of great value in most miRNA sequences because it has been identified as the only miRNA over-expressed in 11 types of solid tumors, such as stomach, prostate, head and neck, breast and so on (Wang et al., 2010). These studies suggest that different miRNA expression patters are connected with various tumor types (Liu et al., 2016a). Therefore, miRNAs can act as effective biomarkers in diagnosing and monitoring human diseases. Nowadays, despite effective technological treatments and the remarkable progress of armarium have been made for human serious diseases, disease is still a major cause of mortality in the modern world. On account of the intimate connection between miRNAs and human healthy, it's imperative to develop reliable and inexpensive devices that enable direct, high sensitive and rapid detection for miRNAs (Ali et al., 2016).

614

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

localized two split segments of ssDNA. Subsequently, these co-localized ssDNA sequences further acted as triggers to initiate CHA to form numerous dsDNA strands, resulting in the recovery of the fluorescent emissions. This approach took the advantage of the signal-amplification features of both HCR and CHA, enabling complete enzyme labelfree detection of miR-141 down to 0.3 fM. Apart from MB, hemin also is used as electrochemical signal enhancer in HCR. In the study of Xiang et al. (2014), target recycling was achieved by the formation of RNA-capture probe (CP) complex on the electrode surface and then release of miRNA from RNA-CP complex (Fig. 2D). Subsequently, the single-strand fragment of CP could initiate HCR, which resulted in the forming hybridization doublestrand DNA structure. Hemin stacked into the G-quadruplex-forming region to form hemin/G-quadruplex and then give a current response. Zhang et al. (2014b) developed a universal ECL biosensor for the detection of miRNA-155 by using hydrogen peroxide (H2O2) as coreactant and hemin as catalyzer. The HCR generated dsDNA polymers gave rise to the intercalation of a lot of hemin which could catalyze the oxidation of H2O2, leading to a remarkably amplified ECL signal output. The biosensor showed a wide linear range from 5 fM to 50 pM and a low detection limit of 1.67 fM. In general, HCR amplification can be widely used because it can bind thousands of auxiliary hairpin probes to amplify electrochemical signal. What's more, HCR procedure is an enzyme-free reaction, which makes it popular in sensing field. Nevertheless, these methods also have several deficiencies, such as laborious labeling techniques or environmentally sensitive, which hinder them from more extensive applications. To address this problem, sensitive and specific electrochemical biosensors for miRNA assay are still needed.

3. Signal-amplification strategy based on oligonucleotide 3.1. Hybridization chain reaction HCR shows great potential in signal amplification for biosensor because it's a kinetics-controlled reaction, possessing unique features such as high sensitivity and good selectivity toward target detection without enzyme. Therefore, many restrictions of enzyme-coupled methods like precise control of pH, temperature and buffer media can be eliminated (Liu et al., 2015a; Zhai et al., 2015). The basic mechanism of HCR was first reported by Pierce and Dirks in 2004 (Dirks et al., 2004). Firstly, the thiolated hairpin capture probe comprising sequences complementary to the target miRNA was assembled on the surface of screen-printed gold electrodes by Au-S bond. In the present of target miRNA, the stem-loop structure of capture probe was unfolded. Then two species of DNA hairpins coupled with the capture probe to trigger the hybridization reaction, therefore produced copolymers to amplify the signal readout. Traditional single strand DNA (T-DNA) probes and sandwich structure DNA probes can hardly detect miRNAs with high sensitivity. So a novel detection probe with a three-dimensional (3D) DNA origami coupling with HCR for miRNA assay was developed, as shown in Fig. 2A (Miao et al., 2015b). The tetrahedral structure could increase accessibility and reactivity, and avoided the introduction of spacer molecules. The sensing part of the unique 3D DNA probe was stemloop structure which was immobilized on the top of 3D DNA probe. The tetrahedron regarded three thiol groups as anchors on gold surface and the thiol groups’ vertices of the tetrahedron facilitated the modification of the tetrahedral DNA on the gold electrode, which reinforced the molecular recognition efficiency. HCR hairpin probe 1 and HCR hairpin probe 2 were then introduced to magnify the electrochemical signal during the detection process. Although this tetrahedral DNA probe showed many advantages, the control of density and orientation of molecules at the electrode interface was still a challenge. From a large quantity of experimental results, a conclusion can be draw that the immobilization of DNA probes on the electrode surface has some intrinsic drawbacks. For example, the immobilization process is complicated and time-consuming. Apart from that, the immobilization procedure may affect the activity of the immobilized biomolecules and the reaction area, which may lead to relatively low binding efficiency. So the label-free and enzyme-free strategies have been widely used in miRNA detection. Recently, methylene blue (MB), an electrochemical indicator, has been introduced to realize such a strategy (Hou et al., 2015). MB can intercalate into double-stranded DNA (dsDNA) through π-π stacking interactions to respond electrical signal (Yuan et al., 2017a). Furthermore, this MB based method exhibited the advantages of simplicity and low cost. HCR are normally coupled with other signal amplification approaches to further enhance the signal response. For example, Wu et al. (2014b) designed an electrochemical biosensor based on HCR and CHA (Fig. 2B). The hairpin-shaped capture probe H1 was first opened by target. In the presence of another hairpin probe H2, hybridization of H1 to H2 began and resulted in the release of target from H1-target complex by strand-displacement reaction. The released target further hybridized with the remaining capture probe H1. After the target recycling process, H1-H2 complex was achieved with an exposed stem of H2. Then, the exposed stem of H2 served as initiator to rigger HCR event, yielding long dsDNA molecule. Ultimately, numerous MB as redox probes intercalated into the minor groove of the long dsDNA polymers to achieve amplified electrochemical signal. This biosensor possessed admirable sensitivity with a detection limit of 3.3 fM and a wide dynamic range with a span of six orders of magnitude. Wei et al. (2016) developed a dual signal-amplification platform based on HCR and CHA (Fig. 2C). The presence of target miRNA triggered the HCR and produced dsDNA polymers, which co-

3.2. Signal-amplification strategy based on nuclease Nuclease is a kind of protein which plays an important role in hydrolyzing phosphate diester bond between nucleotide in the first step of nucleic acid decomposition. According to the different action point of nuclease, it can be classified as exonuclease and endonuclease. Signalamplification techniques based on nuclease have revolutionized the science of biosensor. Samples (DNA or RNA) can be generally amplified thousands or even millions of times in a matter of hours using enzymatic reactions (Zhang et al., 2010). Cyclic enzymatic amplification method (CEAM) and strand displacement amplification (SDA) are the most commonly used techniques. In the CEAM, one target leads to many cycles of target dependent nuclease cleavage of reporter probes for outputting signal amplification (Li et al., 2008; Xiao et al., 2017; Ge et al., 2017). Taking Exonuclease III (Exo III) as an example, Exo III catalyzes the stepwise removal of mononucleotides from the 30-blunt or recessed terminus of duplex DNA. Relying on this unique property, Exo III-based CEAM has been widely used for optical (fluorescence, SPR, UV–Vis, etc.) and electrical signal amplification in detecting DNA, miRNA, proteins and small molecules (Cui et al., 2013). Another commonly used nuclease is DNase I. Cui et al. (2012) developed a DNase I-based CEAM for miRNA detection. However, this assay suffered from high fluorescence background, high detection limit (5 nM) and expensive cost. Meanwhile, T7 exonuclease-based CEAM has also been widely used for miRNA detection, as shown in Fig. 3B (Wang et al., 2014). T7 exonuclease catalyzed the removal of 5′ mononucleotides from the 5′ termini of dsDNA, while its activity on ssDNA was limited. The recycling probe digestion mechanism of the assay leaded to signal amplification and therefore enhanced detection sensitivity of target miRNA. As a result, a detection limit of 0.17 fM was obtained. However, CEAM still suffers from several inherent limitations when used for miRNA detection. For instance, as a linear amplification process, most CEAM can only achieve detection limits of pM level. Furthermore, it is limited by the intrinsic properties of the nucleases

615

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 3. (A) Schematic diagram of T7 exonuclease-based CEAM to detect miRNA. (Reproduced with permission from Wang, et al. (2014). Copyright 2014, American Chemical Society). (B) Schematic representation of utilization of T4 RNA ligase 2 and T7 exonuclease to detecting target miRNA-21. (Reproduced with permission from Li, et al. (2016d). Copyright 2015, Elsevier B.V.). (C) Schematic illustration of miRNA biosensor using KF polymerase and two kinds of exonucleases (Reproduced with permission from Zhu, et al. (2014). Copyright 2014, American Chemical Society).

fication. So KF polymerase gets more application for detecting miRNA (Liu et al., 2016c; Wang et al., 2015; Chi et al., 2017). As displayed in Fig. 3C, Zhu et al. (2014) fabricated a target-triggered recycling SDA reaction with the assistance of KF polymerase. The helper DNA was replaced by the target extension, producing a complete dsDNA with the recognition sites for nicking enzyme. This SDA method allowed for sensitive determination of let-7a miRNA with a detection limit of 0.3 fM. Except KF polymerase, Chen's group (Chen et al., 2016) also used phi29 DNA polymerase (phi29) to design miRNA biosensor which has the abilities of extreme strand displacement and high fidelity continuous polymerization at temperatures between 30 and 40 °C (Alsmadi et al., 2009). Therefore, phi29 has the potential to carry out a SDA reaction with more accurate and comprehensive products, which is promising to improve the sensitivity of detection (Paez et al., 2004). Significantly, a signal-off biosensor for miRNA detection based on a target induced cycling SDA mediated by phi29 polymerase under a mild condition (the physiological temperature) has achieved with homogeneous amplification products and high efficiency. So far, the sensing platform for miRNA which used CEAM and SDA strategies by nuclease has been widely reported. Practices show that the favorable property of nuclease amplification is sufficient to take their advantages of practical based application toward biomedical analysis or clinical diagnosis as long as these nucleases are adequately used, including Exo III, DNase I, T7 exonuclease, T4 RNA ligase, nicking exonuclease, KF polymerase and phi29 polymerase.

used, for that most of the CEAMs are not applicable to RNA targets. Nevertheless, by taking two-stage CEAM, the biosensor can be endowed with excellent sensitivity. Li's group (Li et al., 2016d) employed T4 RNA ligase 2 to initiate specific ligation reaction, and then T7 exonuclease was used to catalyze the first stage CEAM specifically. The second stage of CEAM further amplified the response (Fig. 3B). This two-stage CEAM strategy showed detection limit for miRNA as low as 0.36 fM with remarkable specificity. Additionally, most reactions happened in the solution which was more efficient, and only the last step took place on the electrode surface. This strategy spent much less time and was very convenient for point-of-care testing. Another nuclease based signal-amplification method SDA also has drawn public attention for miRNA detection on account of the advantages of high efficiency, adaptability and simple operation. SDA can provide exponential amplification of a trace of DNA or RNA, which is an important step for nucleic acid detection (Connolly et al., 2011; Wang et al., 2017a; Zhang et al., 2014a;). It has been widely used for the detection of miRNA (Hu et al., 2017; Zhang et al., 2015a; Chen et al., 2017). SDA is a nicking endonuclease-assisted isothermal polymerization reaction activated by specific primer and creates products of single-stranded DNA. Commonly, the polymerase used in SDA is Bst 2.0 DNA polymerase or Klenow fragment (KF) polymerase. However, Bst 2.0 DNA polymerase requires a high temperature of 60–72 °C for optimum enzymatic activity, which may introduce a large amount of byproducts of the non-target fragments and therefore decrease the efficiency of the signal ampli616

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 4C, Wu et al. (2016b) synthesized Pd@HRP label and combined it with target-induced assembly of DNAzyme to develop ultrasensitive electrochemical miRNA biosensor. The composite showed a good redox activity of HRP and electrocatalytic activity toward H2O2. Meanwhile, the binding of RCA-amplified ssDNA concatamer with abundant Pd@HRPDNA probes ensured a strong electrochemical signal output toward target miRNA. Hence, the proposed bioassay exhibited a low detection of 0.2 fM limit with linear range from 3 fM to 1 nM. Recognition mode plays significant roles in specific detection of miRNA with RCA. Currently, most reported recognition modes are direct hybridization-based recognized mode (Xi et al., 2014) and ligation reaction-based recognized mode (Zhang et al., 2014c). Although certain advances toward specific detection of miRNA have been achieved, some limitations still remain in these strategies. Direct hybridization-based recognized mode is based on simple Watson-Crick base pairing, but the thermodynamic energies changes are rather small, which makes this recognized mode suffer from poor specificity (Li et al., 2002). Ligation reaction-based recognized mode is based on the ligation of probe by ligase at perfectly matched terminals. However, when mismatched bases are far from the ligation site, this mode can not present satisfactory specificity. Hence, there is still an urgent need for developing new recognition mode with high specificity to discriminate miRNA from high-homology sequences, whose mismatched bases located in varied positions. Recently, a split recognition mode combined with RCA was developed for highly specific and sensitive detection of miRNA (Wang et al., 2016a) (Fig. 4D). The split recognition mode possessed two specific recognition processes, which were based on toehold-mediated strand displacement reaction (TSDR) and direct hybridization reaction. Two recognition probes, hairpin probe with overhanging toehold domain and assistant probe (AP) were specially designed. Firstly, the toehold domain of hairpin probe and AP recognized part of miRNA simultaneously, and accompanied with TSDR to unfold the hairpin probe and form the stable DNA Y-shaped junction structure (YJS). Next, the AP in YJS could further act as primer to initiate the SDA reactions, which released a great number of trigger sequences. Finally, the trigger sequences hybridized with the padlock DNA which was specially designed with G-quadruplex complementary sequence and part of nicking site for Nt. BbvCI nicking endonuclease cleavage to initiate the circular RCA reaction and produce numerous G-quadruplex sequences. The G-quadruplex sequences could bind with N-methyl mesoporphyrin IX to generate enhanced fluorescence responses. This RCA based method showed high sensitivity toward let-7b with a detection limit of 3.2 pM In the wake of developments of single amplification, dual amplification strategies have caught the scientists attentions due to it can integrate the merits of one fold strategy with high effectiveness. For example, Cui et al. (2014) developed a RCA-CEAM dual amplification method for detection of miRNA. The method achieved a detection limit as low as 12 fM and was capable of distinguishing the single-base mismatch miRNA. Zhang et. al. (2015a) (Fig. 4E) demonstrated a highly sensitive off-on switching ECL sensor for miRNA by coupling the DNAzyme-assisted target recycling and RCA. The first amplification was started by Pb2+ induced target recycling; hence numerous cleaved DNA fragments were generated that acted as RCA primers to trigger the further amplification. RCA combined with other amplification techniques can remarkably enhance the electrochemical signal, and showed outstanding reproducibility, stability and selectivity. In a word, RCA strategy has played key role in miRNA detection because it showed high specificity and greatly amplified the readout signal, and it can incorporate with various functional nucleases to operate the further amplification.

3.3. Rolling circle amplification strategy RCA is an isothermal amplification approach involving numerous copies of the complementary sequence toward original circular template with tandem periodic oligonucleotides to produce a long ssDNA (Dean et al., 2001; Han et al., 2015; Yu et al., 2017). RCA is generally considered as highly sensitive and specific technology resulting from the RCA-synthesized DNA concatemers can bind more signal probes. As an isothermal DNA amplification technique, RCA has been demonstrated to be a versatile tool for miRNA detection. In 2006, Jonstrup et al. developed a method by using miRNA as template to cycle padlock probes and subsequently as a primer for RCA. Linearly amplified detection of miRNA was achieved with a detection limit of 10 pM (Jonstrup et al., 2006). Although RCA is proven to be simple and reliable, the limited sensitivity can not satisfy the requirement of miRNA detection in real samples. On the strength of this scruple, tetrahedral DNA architectures have been attracted intense interest in the field of amplification because of their proven mechanical rigidity and structural stability, as well as the provided effective scaffolds for anchoring of a variety of targets (Bu et al., 2011; Pei et al., 2013). In terms of the distinct properties of tetrahedral DNA structure, researchers not only applied it to HCR, but also combined it with RCA. Miao et al. (2015c) has fabricated an electrochemical method for detection of miRNA (Fig. 4A). Generally, a DNA tetrahedron decorated gold electrode was employed as the recognition probe, and then hybridizations between DNA tetrahedron, target miRNA and primer initiated RCA on the electrode surface. Silver nanoparticles attached to the RCA products provide significant electrochemical signals and a limit of detection as low as 50 aM was achieved. Moreover, homology miRNA family members with only one or two mismatches can be successfully distinguished. Therefore, this method reveals great advancements toward improved sensitivity of RCA. Compared to end-point detection methods, the real-time monitoring makes RCA more suitable for biosensing applications. Some strategies have been developed for the real-time monitoring of RCA reactions (Jiang et al., 2015). It is a task which involves many difficulties during RCA process. As example, SYBR Green and molecular beacons were considered as potential probes for real-time monitoring of RCA (Morrison et al., 1998). However, SYBR Green is not a good choice for real-time RCA since it is a dsDNA fluorescent probe but RCA generates single-stranded products. One drawback to the molecular beacon method is a high background signal due to exonucleolytic degradation. Although Nilsson et al. reported that the use of a molecular beacon composed of 2′-O-Me-RNA residues could address this issue, such a modification would certainly greatly increase the synthetic cost of the probe (Nilsson et al., 2002). As shown in Fig. 4B, Jiang et al. (2016) provided a new approach for applying realtime RCA in biosensing via a specific fluorescent response of commercially available thioflavin T (ThT) to RCA products which contained Gquadruplex structures to recognize miRNA let-7a fast. The proposed method was cost-efficient since no labeled oligonucleotides were used and the dye ThT was inexpensive. Due to the excellent recognition of ThT toward G-quadruplex, low background could be given as long as RCA templates, primers and undesired products couldn’t fold into Gquadruplexes. Such a real-time monitoring strategy was demonstrated to work well for not only traditional linear RCA but also nicking exonuclease-mediated exponential RCA. Unfortunately, with RCA involving multiple monofunctional tags in reported signal probe, RCA presents a risk for sample contamination and causes the complexity and limited efficiency of bioassay. Redox enzyme such as horseradish peroxidase (HRP) is the most promising candidate for resolving the above issues, considering its intrinsic redox center and biocatalytic activity. Compared with a free enzyme, highly loaded enzyme structure is preferable to the construction of high-efficiency biosensor owing to the positive effect of the quantity of enzyme on its amplification efficiency and direct electrochemistry (Das et al., 2007). As displayed in

3.4. Catalyzed hairpin assembly amplification strategy CHA is developed from DNA nanostructure organization, and has received particular interest owing to its excellent property of enzyme617

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 4. (A) Scheme of the tridimensional probe guided RCA-based miRNA assay. (Reproduced with permission from Miao, et al. (2015c). Copyright 2015, American Chemical Society). (B) Working mechanism of the real-time monitoring miRNA let-7a by using the G-quadruplex probe. (Reproduced with permission from Jiang, et al. (2016). Copyright 2016, Elsevier B.V.). (C) Illustration of the miRNA sensing based on Pd@HRP and DNAzyme-aided RCA reaction. (Reproduced with permission from Wu, et al. (2016b). Copyright 2016, Elsevier B.V.). (D) Scheme illustration of miRNA detection strategy by SDA and RCA. (Reproduced with permission from Wang, et al. (2016a). Copyright 2016, Elsevier B.V.). (E) Schematic diagram of miRNA biosensor based on DNAzyme-assisted target recycling and RCA. (Reproduced with permission from Zhang, et al. (2015a). Copyright 2015, American Chemical Society).

mance (Bhadra et al., 2014; Jiang et al., 2014). Aiming at improving the electrochemical properties of CHA-based biosensing strategies for miRNA detection, Zhang et al. (2015b) introduced mismatched base pairs into the breathing sites of the hairpin substrates which could reduce nonspecific CHA products in the absence of target (Fig. 5A). To investigate the effects of the four breath sites on the background signal, two consecutive mismatched base pairs at the end of the stem and the adjacent region were introduced (Fig. 5B). The results displayed that H2D2M2 was the best mismatched site. Compared with the traditional CHA, this kind of mismatched CHA amplification fully decreased the background reaction signal, which improved the sensitivity for detection of target miRNA down to 0.6 pM within 1.5 h. Whereafter, Li et al.

free signal amplification (Liu et al., 2017b; Li et al., 2012; Zhang et al., 2017). A pair of hairpins are triggered by the target miRNA to form a duplex, which lead to the cyclic reuse of the target miRNA and the CHA products without the assistance of enzyme. What's more, hundred-fold catalytic amplification can be achieved by CHA reaction. Recently, miscellaneous CHAs have been developed based on different detection technologies, such as fluorescence, colorimetry and electrochemistry (Liu et al., 2016b; Wu et al., 2016a; Shi et al., 2017). Although CHA shows many advantages, a big background signal, caused by the nonspecific CHA products in the absence of target, is the most trouble, which may counteract the specificity of CHA-based signal amplification methodologies and compromise their analytical perfor618

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 5. (A) Schematic representation of miRNA electrochemical biosensor based on mismatched CHA amplification; (B) Four mismatch positions of H1 and H2. (Reproduced with permission from Zhang et al., 2015b. Copyright 2015, Elsevier B.V.); (C) Schematic diagram of miRNA SPR biosensor based on mismatched CHA. (Reproduced with permission from Li, et al. (2016). Copyright 2016, Elsevier B.V.); (D) QDs-labeled strip miRNA biosensing assay based on CHA. (Reproduced with permission from Deng et al., 2016. Copyright 2016, Elsevier B.V.).

capture probes on the sensor chip, and the massive activated streptavidin-aptamers could capture the streptavidin to achieve enhancement and output of the detection signal. Benefiting from the outstanding performance of the enzyme-free CHA amplification and non-label SPR biosensor, the established biosensor exhibited simple process, high sensitivity and good selectivity. Point-of-care (POC) diagnostics can provide equipment-free, sensitive and robust diagnosis approach for accident points, emergency situations or somewhere low-resource settings, so it has attracted increasing interests recently (Gubala et al., 2012; Hartman et al., 2013; Peeling et al., 2010). Deng et al. (2016) proposed a robust and userfriendly quantum dots (QDs)-labeled strip biosensing platform coupled with CHA for visual, rapid and sensitive detection of miRNA, whereby taking full advantages of the unique optical properties of QDs and convenient performance of target-recycled nonenzymatic amplification strategy. The experimental principle is presented in Fig. 5D. In the system, QDs were served as bright and photostable signal labels for

(2016c) developed the mismatched CHA integrated with SPR biosensor for the highly sensitive detection of miRNA. As shown in Fig. 5C, the SPR biosensing methodology for miRNA-21 was fabricated based on cascade signal amplification of multi-component nucleic acid enzymemediated mismatched CHA (MNAzyme-CHA) strategy. The topical MNAzymes co-recognized the target to form a steady active MNAzyme, which continued to digest multiple hairpin H0 substrates, concomitantly generating a lot of fragments. The H0 fragments could initiate the mismatched CHA cycles and form hairpin H1–H2 complexes which were attached to the sensor surface by streptavidin, leading to a meaningfully amplified SPR signal readout. In the same year, Li et al. (2016a) proposed an enzyme-free SPR biosensor for real-time detecting miRNA based on allosteric effect of mismatched CHA. The presence of target miRNA triggered the allosteric effect of CHA amplification, which brought about the recycling of the target miRNA and produced large amounts of CHA products and activated streptavidin-aptamers. Meanwhile, the plentiful CHA products could hybridize with the 619

620

Combining padlock RCA with CoFe2O4 MNPs by nanoelectrocatalysis Toehold-mediated SDA reaction formed YJS coupling with RCA G-quadruplex probe-mediated real-time exponential RCA In situ formation of molybdophosphate nanosphere combining with CHA reaction Allosteric effect of mismatched CHA for enzyme-free SPR Nonenzymatic CHA for QDs-labeled strip biosensor

Phi 29 polymerase synergic target-cycling and RCA

Two-stage CEAM by T4 RNA ligase 2 and T7 exonuclease SDA coupled with TWJ

LOD: limit of detection; MNPs: magnetic nanoparticles; TWJ: three-way junction.

CHA

RCA

Nuclease amplification

5 pM-100 nM 2–200 fM

1 pM 200 aM

10 aM-1 nM

4 aM 10 fM-1 nM

10 pM-10 nM

3.2 pM

1.64 fM

1 fM-2 nM

0.3 fM

100 aM-100 pM

1 fM-1 nM

0.68 fM 22 aM

100 pM−1 fM

1–100 pM

1 fM-1 nM

10–16−10−10 M

Linear range

0.36 fM

0.44 fM, 0.46 fM 31.8 fM

Multifunctional MNPs probe coupling with HCR

HRP-assisted HCR and enzyme-assisted visualization

2 aM

T-DNA probe combining with AuNPs triggered HCR

HCR

LOD

Mechanism

Amplification techniques

Table 1 Comparison of different amplification strategies based on oligonucleotide for miRNA detection.

Broad linear range, high specificity, simple operation, cost-efficient Application in two cancerous, nanomaterial for amplifying signal Enzyme-free, simple process, good repeatability Simple and convenient operation, rapid and sensitive analysis

Target for cyclic utilization, broad linear range, high sensitivity High specificity, exerting the merits of CoFe2O4 MNPs Two split recognition mode

Quick, simple, broad linear range

Rapid and convenient

High sensitivity, easy operation, nonenzymatic detection Simultaneous detection, exerting the merits of MNPs High sensitivity, high specificity

Advantage

Narrow linear range

Low throughput, requires sophisticated read-out system No multiplex potential due to complicated preparation of the sensor Low sensitivity

Low sensitivity

The reaction time of immobilization step is too long, special enzymes needed Unknown efficiency and reliability of the conjugation of TWJ to electrode surface More condition optimization needed, special enzymes needed Complicated procedure of previous preparation

Narrow linear range

Complicated nucleic acid chain design

Complicated design of T-DNA, non-quantitative

Limitation

(Deng et al., 2016)

(Li et al., 2016a)

(Wang et al., 2016a) (Jiang et al., 2016) (Cai et al., 2017)

(Chen et al., 2016) (Yu et al., 2017)

(Hu et al., 2017)

(Miao et al., 2015b) (Yuan et al., 2017b) (Ying et al., 2017) (Li et al., 2016d)

Ref.

Y.-X. Chen et al.

Biosensors and Bioelectronics 99 (2018) 612–624

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

POC, which endowed the biosensor good detection efficiency. CHA was used to amplify detection signal. For miRNA detection, this assay was 10-fold more sensitive than that with a conventional colloidal goldbased test strip. CHA strategy is a simple and nimble method which can be ascribed to its enzyme-free peculiarity that needn’t consider the optimal temperature of enzyme. Therefore it has attracted great interest in miRNA detection. In addition, the target miRNA can be cyclic utilized for many times during CHA procedure, so that the designed biosensor achieves excellent sensitivity even though the level of target miRNA is tiny. In addition, the operation of CHA step is one-pot that operated in an additional vessel rather than adding solution to the working electrode stage by stage, which can reduce the relative error. The analytical performance of the various amplification techniques based on oligonucleotides for miRNA detection is summarized in Table 1.

specific detection of miRNA-21 was realized. Most important, this method was resistant to matrix interference, even in complex biofluids like serum. Tang et al. reported antibody-coated AuNPs can improve the sensitivity of electrochemical immunoassays using p-aminophenol (p-AP) redox cycling (Tang et al., 2011). Liu et al. (2014) attempted to use a triple-signal amplification (AuNPs, ALP and p-AP redox cycling) to improve the performance of miRNA biosensor. As a result, the increased linearly with the miRNAs concentration over a range of 10 fM-5 pM and a detection limit of 3 fM were achieved. The results indicate that the multifunctional AuNPs are extremely stable and can be used to recognize complicated miRNAs sequences through the formation of boronate ester covalent bonds, even obviating the use of specific DNA/RNA modification for molecular recognition and signal amplification, reducing the operation complexity and the cost.

4. Signal amplification based on nanomaterials

Researchist gradually finds carbon materials have better properties than other materials in hardness, optical properties, heat resistance, electrical conductivity, surface and interface properties. Recently, carbon nanomaterials have been widely applied in the detection of miRNA. Frequently-used carbon nanomaterials are graphene and carbon nanotubes (CNTs). Graphene is a monolayer of sp2-bonded carbon atoms arranged in honeycomb lattice, which is a typical two-dimension carbon nanomaterial (Mas-Balleste et al., 2011; Huang et al., 2014c). Since graphene was discovered in 2004, it has attracted extensive attention because of its unique and remarkable properties, such as excellent electrical conductivity, large theoretical specific surface area and strong mechanical strength (Bo et al., 2017; Huang et al., 2014d). In recent years, an increasing number of graphene-based miRNA sensors have been reported (Dong et al., 2012). Meanwhile, graphene oxide (GO) also is a widely used electrode material, whose surface possesses many carboxylic acid and hydroxyl groups (Gao et al., 2017), making it more water-soluble and suitable for biochemical and biomedical analysis (Huang et al., 2015). Many biosensors have utilized GO to establish sensing platform for the detection of miRNA. In addition, graphene quantum dots (GQDs) also are used to fabricate ECL miRNA biosensors. It's a zero-dimensional nanomaterial obtained from graphene which have characteristics of large surface area, good surface grafting by π-π conjugated network (Lin et al., 2014; Zhang et al., 2016). Zhang et al. (2015c) introduced a substrate to load GQDs through π-π stacking, realizing the solid-state GQDs applied in miRNA detection. CNTs stimulate global research activity due to their ultra-high specific surface area and outstanding electrical, mechanical and electrochemical properties (Huang et al., 2014e). They can be divided into single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs). In general, SWCNT is a single molecular nanomaterial, which is formed of only a layer that rolls a single sheet of graphite into a seamless molecular cylinder, while MWCNT is composed of more than two layers of curly graphite sheet, and the distance between each layer is approximately 0.42 nm (Popov et al., 2004). Comparing with the most of the commercially available miRNA sensors that based usually on metal oxides, silicon and other materials, the CNT-based miRNA biosensors have several advantages. First, with large surface area ratio and hollow pipe, CNTs can be used to immobilize enzyme with high biological activity remained (Jacobs et al., 2010). Next, CNTs have an outstanding ability to mediate fast electron-transfer kinetics hence promote the electron-transfer reactions. The last but not least, CNTs's structure shows high stability due to their large intertube attraction energy. These particular characteristics have inspired the increasing research interest in the applications of CNTs as components for high-performance miRNA biosensors. As displayed in Fig. 6B, Liu et al. (2015a) proposed an electrochemical assay for detecting miRNA-21. A three-dimensional (3D) layer-by-layer nanostructure composed of oxidized SWCNTs (SWCNTs-ox) and

4.2. Carbon nanomaterials

As more and more signal-amplification techniques become available at low cost, the use of nano-sized materials for biosensor development is attractive. Most recently, nanomaterials are widely used to detect miRNA due to their high surface-to-volume ratio and size-dependent properties. In general, for constructing sensing platforms, nanomaterials usual are served as electrode supporting substrates, carriers for signal elements, tracers based on their direct electrochemistry, separators and collectors or catalysts and mediators to regulate the electron transfer process (Wu et al., 2014a). In this work, Au nanoparticles (AuNPs), carbon nanomaterials and transitionmetal dichalcogenides (TMDCs) used in biosensor for miRNA detection are mainly reviewed. 4.1. AuNPs AuNPs, like other nanostructured materials, provide a suitable micro-environment for the immobilization of probe, aptamer and multifarious enzymes. Their presence in miRNA biosensors is crucial for loading significant component for signal amplification and enabling electronic communication between the redox center of enzymes and the electrode substrate (Madalina et al., 2016). AuNPs possess high stability, conductivity, biocompatibility and size-related electronic, magnetic and optical properties and have been widely applied for miRNA detection. AuNPs have large surface to volume ratios and are biocompatible, which make them very suitable to be used as carriers for biological molecules (Huang et al., 2015a, b, 2014b). Peng et al. (2014) used streptavidin-AuNPs (SA-AuNPs), alkaline phosphatase (ALP) and polymerase extension as the signal-amplification elements to detect miRNA. This biosensor showed a detection limit of 99.2 fM toward target miRNA. Of course, AuNPs can not only act as carriers but also play important roles in the fabrication of sensing substrate to bind capture probe though Au-S bond. Yin. et al. (2012) developed an “offon” electrochemical sensing platform for miRNA, which combined two functions of AuNPs (Fig. 6A). In the work, graphene and AuNPs were used as supporting substrate of sensor. Two kinds of AuNPs conjugates that prepared with different report DNAs were introduced into biosensor for increasing the amounts of the biotin groups. The detection sensitivity of the miRNA biosensor was then improved after abundant streptavidin-HRP conjugates were immobilized to the biosensor by biotin-streptavidin reaction. The linear range of the biosensor was 0.01–700 pM and the detection limit was 6 fM. Wang et. al (2016b) developed an enzyme-free SPR biosensor for miRNA detection based on AuNPs and DNA supersandwich. The DNA-linked AuNPs not only hybridized with the capture DNA on the Au film to amplify SPR signal but also initiated the subsequent secondary amplification. Due to the dual amplification of AuNPs and supersandwich, sensitive and 621

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Fig. 6. (A) The electrochemical detection of miRNA through three steps of amplification by AuNPs. (Reproduced with permission from Yin, et al. (2012). Copyright 2012, The Royal Society of Chemistry); (B) Schematic illustrations of the electrochemical miRNA biosensor based on SWCNTs-ox. (Reproduced with permission from Liu, et al. (2015a). Copyright 2015, Elsevier B.V.); (C) Schematic illustration of the strategy for miRNA detection based on MoS2; (D) The SEM image of MoS2 based biosensor. (Reproduced with permission from Shuai, et al. (2017a). Copyright 2017, Elsevier B.V.).

for 2D materials to be used in the next generation switching and optoelectronic devices. Similarly to graphene, the present methods for TMDCs preparation may be classified as top-down methods and bottom-up methods. Top-down methods are aimed at changing bulky and layered compounds into single- and few-layer 2D TMDCs including mechanical exfoliation, liquid exfoliation and chemical or electrochemical exfoliation. Bottom-up methods mainly include chemical vapor deposition (CVD) and solvothermal or hydrothermal methods, which are simple, scalable and readily controlled. In contrast to graphene, TMDCs nanosheets can be facilely synthesized in large scale and directly dispersed in aqueous solution without the need of surfactants or oxidation treatment (Xi et al., 2014). Until now, many electrochemical sensing platforms have been fabricated based on TMDCs due to their distinguished properties (Huang et al., 2016a, 2015c, 2015d, 2014f; Liu et al., 2016d, 2015c). MoS2 is a typical family member of TMDCs. It is composed of Mo metal layers sandwiched between two sulfur layers and stacked together by weak Van der Waals interactions (Huang et al., 2014a, g). The layered structure of MoS2 is expected to act as an excellent functional material because the 2D electron-electron correlations among Mo atoms would aid in enhancing planar electric transportation properties (Su et al., 2017). Based on the outstanding performance of MoS2, Huang's group have successfully applied it into electrochemical

nanodiamonds (NDs). It expanded the electrode surface areas and promoted electron transfer. The proposed miRNA biosensor exhibited good reproducibility and stability, as well as high sensitivity with the linear range of 10 fM-1.0 nM and the detection limit of 1.95 fM. 4.3. Transition-metal dichalcogenides Transition-metal dichalcogenides (TMDCs) are an interesting family of 2D materials with an X-M-X layered structure, such as WS2, MoS2, SnS2 and VS2. These materials form sandwich structures with the chalcogen atoms in two hexagonal planes separated by a plane of metal atoms (Wang et al., 2017b). TMDCs as 2D layered nanomaterials analogous to graphene have attracted more and more attentions in the field of electrochemistry due to their large surface specific area, high electronic conductivity, eminent catalytic properties and fast heterogeneous electron transfer that depend on their composition and layer thickness ranging from semiconductors and semimetals to true superconductors and metals (Huang et al., 2016b; Shuai et al., 2017b, c, 2016b, ; Liu et al., 2016b). The process or conversion can be realized by modifying the electron filling by doping, quantum size effect, field effect, or intercalation of molecules, atoms, and ions into Van der Waals layers. This property of TMDCs is inspiring, which will largely compensate the weakness of gapless graphene, thus making it possible 622

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

sensing field (Fang et al., 2015; Huang et al., 2015e, 2014h, i). For instance, they prepared hollow MoS2 microcubes as electrode material coupled with duplex-specific nuclease and enzyme signal amplification for miRNA-21 detection in human serum samples, as shown in Fig. 6C (Shuai et al., 2017a). Herein, MoS2 microcubes with large specific surface coupling with AuNPs act as an excellent sensing substrate to immobilize more capture probe for signal amplification. Fig. 6D exhibits the SEM image of biosensor after reacted with miRNA. The layered MoS2 sheets formed microcubes and the AuNPs equably distribute in the MoS2 microcubes can be observed easily. The results revealed that signal readout greatly increased when MoS2 microcubes were employed in sensor construction, indicating MoS2 microcubes played an amplifying signal role in the sensor.

References Alan, K.L.T., Cai, L.L., Gao, Z.Q., 2014. Electrochim. Acta 126, 19–30. Ali, M., Behzad, M., Behzad, B., 2016. Biomed. Pharmacother. 84, 705–713. Alsmadi, O., Alkayal, F., Monies, D., Meyer, B.F., 2009. BMC Res. Notes 2, 48. Asahiro, M., Tsutomu, M., 2015. Hepatol. Res. 45, 128–141. Ayaz, L., Görür, A., Yaroglu, H.Y., Özcan, C., Tamer, L., 2013. J. Cancer Res. Clin. Oncol. 139, 1499–1506. Baldán, Á., Fernández-Hernando, C., 2016. Curr. Opin. Lipidol. 27, 623–629. Bhadra, S., Ellington, A.D., 2014. Nucleic Acids Res. 42, e58. Bo, X.J., Zhou, M., Guo, L.P., 2017. Biosens. Bioelectron. 89, 167–186. Bu, N.N., Tang, C.X., He, X.W., 2011. Chem. Commun. 47, 7689–7691. Cai, W., Xie, S.B., Tang, Y., Chai, Y.Q., Yuan, R., Zhang, J., 2017. Talanta 163, 65–71. Chen, A.Y., Ma, S.Y., Zhuo, Y., Chai, Y.Q., Yuan, R., 2016. Anal. Chem. 88, 3203–3210. Chen, Y.X., Huang, K.J., Lin, F., Fang, L.X., 2017. Talanta 175, 168–176. Chi, B.Z., Liang, R.P., Qiu, W.B., Yuan, Y.H., Qiu, J.D., 2017. Biosens. Bioelectron. 87, 216–221. Clancy, E., Burke, M., Arabkari, V., Barry, T., Kelly, H., Dwyer, R.M., Kerin, M.J., Smith, T.J., 2017. Anal. Bioanal. Chem. 409, 3497–3505. Connolly, A.R., Trau, M., 2011. Nat. Protoc. 6, 772–778. Cui, L., Lin, X., Lin, N., Song, Y., Zhu, Z., Chen, X., Yang, C.J., 2012. Chem. Commun. 48, 194–196. Cui, L., Ke, G., Lin, X., Song, Y., Zhang, H., Guan, Z., Zhu, Z., Yang, C.J., 2013. Methods 63, 202–211. Cui, L., Zhu, Z., Lin, N.H., Zhang, H.M., Guan, Z.C., Yang, C.Y., 2014. Chem. Commun. 50, 1576–1578. Das, J., Jo, K., Lee, J.W., Yang, H., 2007. Anal. Chem. 79, 2790–2796. Dean, F.B., Nelson, J.R., Giesler, T.L., Lasken, R.S., 2001. Genome Res. 11, 1095–1099. Deng, H.P., Liu, Q.W., Wang, X., Huang, R., Liu, H.X., Lin, Q.M., Zhou, X.M., Xing, D., 2016. Biosens. Bioelectron. 87, 931–940. Dirks, R.M., Pierce, N.A., 2004. P. Natl. Acad. Sci. USA 101, 15275–15278. Dong, H.F., Zhang, J., Ju, H.X., Lu, H.T., Wang, S.Y., Jin, S., Hao, K.H., Du, H.W., Zhang, X.J., 2012. Anal. Chem. 84, 4587–4593. Fang, L.X., Huang, K.J., Liu, Y., 2015. Biosens. Bioelectron. 71, 171–178. Fuentes-Mattei, E., Giza, D.E., Shimizu, M., Ivan, C., Manning, J.T., Tudor, S., Ciccone, M., Kargin, O.A., Zhang, X., Mur, P., Amaral, N.S., Chen, M., Tarrand, J.J., Lupu, F., Ferrajoli, A., Keating, M.J., Vasilescu, C., Yeung, S.C.J., Calin, G.A., 2017. EBioMed 20, 182–192. Gao, Y.P., Wu, X., Huang, K.J., Xing, L.L., Zhang, Y.Y., Liu, L., 2017. CrystEngComm 19, 404–418. Ge, J., Geng, X., Du, Y.H., Chen, J.J., Zhang, L., Bai, D.M., Ji, D.Y., Hu, Y.L., Li, Z.H., 2017. Sens. Actuators B-Chem. 249, 189–194. Gramantieri, L., Ferracin, M., Fornari, F., 2007. Cancer Res. 67, 6092–6099. Gubala, V., Harris, L.F., Ricco, A.J., Tan, M.X., Williams, D.E., 2012. Anal. Chem. 84, 487–515. Gupta, S.K., Foinquinos, A., Thum, S., Remke, J., Zimmer, K., Bauters, C., Groote, P., Boon, R.A., Windt, L.J., Preissl, S., Hein, L., Batkai, S., Pinet, F., Thum, T., 2016. J. Am. Coll. Cardiol. 68, 1557–1571. Han, D., Wu, C., You, M., Zhang, T., Wan, S., Chen, T., Qiu, L., Zheng, Z., Liang, H., Tan, W., 2015. Nat. Chem. 7, 835–841. Harmanci, D., Erkan, E.P., Kocak, A., Akdogan, G.G., 2017. Biomed. Rep. 6, 599–604. Hartman, M.R., Ruiz, R.C., Hamada, S., Xu, C., Yancey, K.G., Yu, Y., Han, W., Luo, D., 2013. Nanoscale 5, 10141–10154. Homola, J., 2008. Chem. Rev. 108, 462–493. Hong, C.Y., Chen, X., Liu, T., Li, J., Yang, H.H., Chen, J.H., Chen, G.N., 2013. Biosens. Bioelectron. 50, 132–136. Hou, T., Li, W., Liu, X.J., Li, F., 2015. Anal. Chem. 87, 11368–11374. Hu, R., Wang, G.L., Yuan, R., Xu, Y.J., Yu, T.X., Zhong, L., Zhou, Q., Ding, S.J., 2017. J. Electroanal. Chem. 789, 160–166. Huang, K.J., Liu, Y.J., Wang, H.B., Wang, Y.Y., Liu, Y.M., 2014a. Biosens. Bioelectron. 55, 195–202. Huang, K.J., Liu, Y.J., Shi, G.W., Yang, X.R., Liu, Y.M., 2014b. Sens. Actuators B-Chem. 201, 579–585. Huang, K.J., Liu, Y.J., Cao, J.T., Wang, H.B., 2014c. RSC Adv. 4, 36742–36748. Huang, K.J., Liu, Y.J., Wang, H.B., Gan, T., Liu, Y.M., Wang, L.L., 2014d. Sens. Actuators B-Chem. 191, 828–836. Huang, K.J., Zhang, J.Z., Xing, K., 2014e. Electrochim. Acta 149, 28–33. Huang, K.J., Liu, Y.J., Shi, G.W., Yang, X.R., Liu, Y.M., 2014f. Sens. Actuators B-Chem. 201, 579–585. Huang, K.J., Wang, L., Zhang, J.Z., Wang, L.L., Mo, Y.P., 2014g. Energy 67, 234–240. Huang, K.J., Liu, Y.J., Liu, Y.M., Wang, L.L., 2014h. J. Hazard. Mater. 276, 207–215. Huang, K.J., Zhang, J.Z., Liu, Y.J., Wang, L.L., 2014i. Sens. Actuators B-Chem. 194, 303–310. Huang, K.J., Liu, Y.J., Zhang, J.Z., 2015a. Microchim. Acta 182, 409–417. Huang, K.J., Liu, Y.J., Zhang, J.Z., Liu, Y.M., 2015b. Sens. Actuators B-Chem. 209, 570–578. Huang, K.J., Liu, Y.J., Zhang, J.Z., Cao, J.T., Liu, Y.M., 2015c. Biosens. Bioelectron. 67, 184–191. Huang, K.J., Liu, Y.J., Zhai, Q.F., 2015d. J. Mater. Chem. B 3, 8180–8187. Huang, K.J., Wang, L., Zhang, J.Z., Xing, K., 2015e. J. Electroanal. Chem. 752, 33–40. Huang, K.J., Shuai, H.L., Chen, Y.X., 2016a. Sens. Actuators B-Chem. 225, 391–397. Huang, K.J., Shuai, H.L., Zhang, J.Z., 2016b. Biosens. Bioelectron. 77, 69–75. Huang, X.H., Wang, Q., Chen, J.S., 2009. Hepatol. Res. 39, 786–794. Jacobs, C.B., Peairs, M.J., Venton, B.J., 2010. Anal. Chim. Acta 662, 105–127.

5. Conclusion and future perspectives With the progress of science and technology, scientists put more emphasis on the precision medicine, and this is an inevitable tendency of current medicine. Since the discovery of miRNAs in 1990s, their close relationship with human healthy suggested that they could be used as biomarkers to improve response unto diseases treatment. As miRNAs play significant roles in human health, the importance of them has become clearer, and the number of biosensors available to analyze various different functional miRNAs has grown. Therefore, looking for convenient, rapid, efficient, ingenious miRNA biosensing platform is extremely urgent. In this work, the development and applications of miRNA are summarized. Furthermore, the progresses about the rapidly developing domain of diverse amplification-based miRNA biosensors in recent years are reviewed. The report highlighted the biological amplification based on oligomeric nucleotides which mainly included HCR, nuclease amplification, RCA, CHA and amplification based on nanomaterials that chiefly contained AuNPs, carbon materials and TMDCs. Finally, the advantages and characteristics as well as defects of the abovementioned miRNA amplification strategies are concluded. It is obvious that single conventional amplified methods have been validated, suggesting that a single universal normalizer for all sample types of miRNA is unlikely to reveal super performance, yet the use of different combination of several amplifications for miRNA sample types may be more appropriate. Furthermore, there are still some important challenges to meet before miRNA diagnosis become a reality. The ideal miRNA profiling technique should be efficient, precise and diversiform which possesses high specificity against other family RNAs, simple and convenient operation, good stability and biocompatibility for the immobilization of the signal probes, and a wide dynamic range from attomolar to nanomolar concentrations. It is still a long way to go for the multifarious amplification techniques to achieve such an ideal goal and overcome false positive, accurate quantification or other problems before they are applied in the clinic. It is firmly believed that with the solution of those challenges, the further sensitive detection of miRNA will come true in the future and this will provide a new perspective for figuring out the pathways of miRNA involving in clinical research. More importantly, this meaningful study of miRNA will make unimaginable contributions to the promotion and development of medical approach. Acknowledgments This work was supported by the National Natural Science Foundation of China (21475115), Program for University Innovative Research Team of Henan (15IRTSTHN001), Henan Provincial Science and technology innovation team (C20150026), Nanhu Scholars Program of XYNU and Henan Science and Technology Cooperation Project (172106000064).

623

Biosensors and Bioelectronics 99 (2018) 612–624

Y.-X. Chen et al.

Shuai, H.L., Huang, K.J., Xing, L.L., Chen, Y.X., 2016a. Biosens. Bioelectron. 86, 337–345. Shuai, H.L., Huang, K.J., Chen, Y.X., 2016b. J. Mater. Chem. B 4, 1186–1196. Shuai, H.L., Huang, K.J., Chen, Y.X., Fang, L.X., Jia, M.P., 2017a. Biosens. Bioelectron. 89, 989–997. Shuai, H.L., Wu, X., Huang, K.J., Zhai, Z.B., 2017b. Biosens. Bioelectron. 94, 616–625. Shuai, H.L., Huang, K.J., Zhang, W.J., Cao, X.Y., Jia, M.P., 2017c. Sens. Actuators BChem. 243, 403–411. Singh, R., Ramasubramanian, B., Kanji, S., Chakraborty, A.R., Haque, S.J., Chakravarti, A., 2016. Cancer Lett. 381, 113–121. Slattery, M.L., Wolff, E., Hoffman, M.D., Pellatt, D.F., Milash, B., Wolff, R.K., 2011. Genes Chromosomes Cancer 50, 196–206. Su, S., Cao, W.F., Liu, W., Lu, Z.W., Zhu, D., Chao, J., Weng, L.X., Wang, L.H., Fan, C.H., Wang, L.H., 2017. Biosens. Bioelectron. 94, 552–559. Tang, J., Tang, D., Su, B., Huang, J., Qiu, B., Chen, G., 2011. Biosens. Bioelectron. 26, 3219–3226. Válóczi, A., Hornyik, C., Varga, N., Burgyán, J., Kauppinen, S., Havelda, Z., 2004. Nucleic Acids Res. 32, e175. Varnholt, H., Drebber, U., Schulze, F., 2008. Hepatology 47, 1223–1232. Wang, F., Zheng, Z., Guo, J., Ding, X., 2010. Gynecol. Oncol. 119, 586–593. Wang, M., Fu, Z.L., Li, B.C., Zhou, Y.L., Yin, H.S., Ai, S.Y., 2014. Anal. Chem. 86, 5606–5610. Wang, M.N., Shen, B., Yuan, R., Cheng, W., Xu, H.B., Ding, S.J., 2015. J. Electroanal. Chem. 756, 147–152. Wang, Q., Liu, R.J., Yang, X.H., Wang, K.M., Zhu, J.Q., He, L.L., Li, Q., 2016b. Sens. Actuators B-Chem. 223, 613–620. Wang, R., Wang, L., Zhao, H.Y., Jiang, W., 2016a. Biosens. Bioelectron. 86, 834–839. Wang, X.C., Liu, W.W., Yin, B.B., Sang, Y.W., Liu, Z.P., Dai, Y., Duan, X.Z., Zhang, G., Ding, S.J., Tao, Z.H., 2017a. Microchim. Acta, 1–8. Wang, Y.H., Huang, K.J., Wu, X., 2017b. Biosens. Bioelectron. 97, 305–316. Wei, Y.L., Zhou, W.J., Li, X., Chai, Y.Q., Yuan, R., Xiang, Y., 2016. Biosens. Bioelectron. 77, 416–420. Wightman, B., Ha, I., Ruvkun, G., 1993. Cell 75, 855–862. Wong, Q.W., Lung, R.W., Law, P.T., 2008. Gastroenterology 135, 257–269. Wu, H., Liu, Y.L., Wang, H.Y., Wu, J., Zhu, F.F., Zou, P., 2016a. Biosens. Bioelectron. 81, 303–308. Wu, L., Xiong, E.H., Zhang, X., Zhang, X.H., Chen, J.H., 2014a. Nano Today 9, 197–211. Wu, X.Y., Chai, Y.Q., Yuan, R., Zhuo, Y., Chen, Y., 2014b. Sens. Actuators B-Chem. 203, 296–302. Wu, Y.M., Sheng, K., Liu, Y.Q., Yu, Q., Ye, B.X., 2016b. Anal. Chim. Acta 948, 1–8. Xi, Q., Zhou, D.M., Kan, Y.Y., Ge, J., Wu, Z.K., Yu, R.Q., Jiang, J.H., 2014. Anal. Chem. 86, 1361–1365. Xiang, G.M., Jiang, D.N., Luo, F.K., Liu, F., Liu, L.L., Pu, X.Y., 2014. Sens. Actuators BChem. 195, 515–519. Xiao, K.Y., Liu, J., Chen, H., Zhang, S., Kong, J.L., 2017. Biosens. Bioelectron. 91, 76–81. Yang, S., Dai, X., Stogin, B.B., 2016. P. Natl. Acad. Sci. 113, 268–273. Yang, X., Mattes, W.B., Shi, Q., Weng, Z.Q., Salminen, W.F., 2014. MicroRNAs Toxicol. Med., 363–387. Yin, H.S., Zhou, Y.L., Chen, C.X., Zhu, L.S., Ai, S.Y., 2012. Analyst 137, 1389–1395. Ying, N., Sun, T.F., Chen, Z.B., Song, G.P., Qi, B.Y., Bu, S.J., Sun, X.W., Wan, J.Y., Li, Z.H., 2017. Anal. Biochem. 528, 7–12. Yu, N., Wang, Z., Wang, C.C., Han, J., Bu, H.Y., 2017. Anal. Chim. Acta 962, 24–31. Yuan, Y.H., Wu, Y.D., Chi, B.Z., Wen, S.H., Liang, R.P., Qiu, J.D., 2017b. Biosens. Bioelectron.. http://dx.doi.org/10.1016/j.bios.2017.06.022. Yuan, Y.L., Liu, G.P., Chai, Y.Q., Yuan, R., 2017a. Sens. Actuators B-Chem. 246, 402–407. Zen, K., Zhang, C.Y., 2012. Med. Res. Rev. 32, 326–348. Zhai, Q., He, Y., Li, X.L., Guo, J., Li, S.Q., Yi, G., 2015. J. Electroanal. Chem. 758, 20–25. Zhang, C., Xing, D., 2010. Chem. Rev. 110, 4910–4947. Zhang, H., Wang, Q., Yang, X.H., Wang, K.M., Li, Q., Li, Z.P., Gao, L., Nie, W.Y., Zheng, Y., 2017. Analyst 142, 389–396. Zhang, L.R., Zhu, G.C., Zhang, C.Y., 2014a. Anal. Chem. 86, 6703–6709. Zhang, P., Wu, X.Y., Chai, Y.Q., Yuan, R., 2014b. Analyst 139, 2748–2753. Zhang, P., Wu, X.Y., Yuan, R., Chai, Y.Q., 2015a. Anal. Chem. 87, 3202–3207. Zhang, P., Zhuo, Y., Chang, Y.Y., Yuan, R., Chai, Y.Q., 2015b. Anal. Chem. 87, 10385–10391. Zhang, P.B., Liu, Y., Zhang, Y.J., Liu, C.H., Wang, Z.B., Li, Z.P., 2013. Chem. Commun. 49, 10013–10015. Zhang, P.B., Zhang, J.Y., Wang, C.L., Liu, C.H., Wang, H., Li, Z.P., 2014c. Anal. Chem. 86, 1076–1082. Zhang, T.T., Zhao, H.M., Fan, G.F., Li, Y.X., Li, L., Quan, X., 2016. Electrochim. Acta 190, 1150–1158. Zhang, Y., Yan, Y.R., Chen, W.H., Cheng, W., Li, S.Q., Ding, X.J., Li, D.D., Wang, H., Huang, X., Ding, S.J., 2015b. Biosens. Bioelectron. 68, 343–349. Zhao, Y.X., Chen, F., Li, Q., Wang, L.H., Fan, C.H., 2015. Chem. Rev. 115, 12491–12545. Zhu, G.C., Liang, L., Zhang, C.Y., 2014. Anal. Chem. 86, 11410–11416. Zhu, H., Leung, S.W., 2015. Diabetologia, 900–911.

Jiang, H.X., Zhao, M.Y., Niu, C.D., Kong, D.M., 2015. Chem. Commun. 51, 16518–16521. Jiang, H.X., Liang, Z.Z., Ma, Y.H., Kong, D.M., Hong, Z.Y., 2016. Anal. Chim. Acta 943, 114–122. Jiang, Y.S., Bhadra, S., Li, B.L., Ellington, A.D., 2014. Angew. Chem. Int. Ed. 126, 1876–1879. Jin, X.F., Wu, N., Wang, L., Li, J., 2013. Cell. Mol. Neurobiol. 33, 601–613. João, C., Elazer, R.E., Natalie, A., 2015. Adv. Drug Deliv. Rev. 81, 169–183. Jonstrup, S.P., Koch, J., Kjems, J., 2006. RNA 12, 1747–1752. Khan, N., Cheng, J., Pezacki, J.P., Berezovski, M.V., 2011. Anal. Chem. 83, 6196–6201. Kung, L.H.W., Zaki, S., Ravi, V., Rowley, L., Smith, M.M., Bell, K.M., Bateman, J.F., Little, C.B., 2017. Osteoarthr. Cartil. 25, 426–434. Leni, M., Kara, E.B., Joanne, T., Jon, W., Clay, B.M., Melissa, P., 2014. J. Cell. Mol. Med. 18, 371–390. Li, B.C., Li, X., Wang, M., Yang, Z.Q., Yin, H.S., Ai, S.Y., 2015. J. Solid State Electr. 19, 1301–1309. Li, B.C., Liu, F., Peng, Y.Y., Zhou, Y.L., Fan, W.X., Yin, H.S., Ai, S.Y., Zhang, X.S., 2016d. Biosens. Bioelectron. 79, 307–312. Li, B.L., Jiang, Y., Chen, X., Ellington, A.D., 2012. J. Am. Chem. Soc. 134, 13918–13921. Li, J.B., Lei, P.H., Ding, S.J., Zhang, Y., Yang, J.R., Cheng, Q., Yan, Y.R., 2016a. Biosens. Bioelectron. 77, 435–441. Li, J.J., Chu, Y., Lee, B.Y., Xie, X.S., 2008. Nucleic Acids Res. 36, 36. Li, J.Y., Yong, T.Y., Michael, M.Z., Gleadle, J.M., 2010. Nephrology 15, 599–608. Li, Q., Luan, G., Guo, Q., Liang, J., 2002. Nucleic Acid. Res. 30, 5. Li, R.D., Yin, B.C., Ye, B.C., 2016b. Biosens. Bioelectron. 86, 1011–1016. Li, X.M., Cheng, W., Li, D.D., Wu, J.L., Ding, X.J., Cheng, Q., Ding, S.J., 2016c. Biosens. Bioelectron. 80, 98–104. Lin, L.P., Rong, M.C., Luo, F., Chen, D.M., Wang, Y., Chen, X., 2014. TrAC Trend Anal. Chem. 54, 83–102. Liu, C., Lv, S.F., Gong, H., Chen, C.Y., Chen, X.M., Cai, C.Q., 2017b. Talanta 174, 336–340. Liu, H.Y., Bei, X.Q., Xia, Q.T., Fu, Y., Zhang, S., 2016a. Microchim. Acta 183, 297–304. Liu, H.Y., Tian, T., Ji, D.D., Ren, N., Ge, S.G., Yan, M., Yu, J.H., 2016b. Biosens. Bioelectron. 85, 909–914. Liu, L., Xia, N., Liu, H.P., Kang, X.J., Liu, X.S., Xue, C., He, X.L., 2014. Biosens. Bioelectron. 53, 399–405. Liu, L.Z., Song, C., Zhang, Z., Yang, J., Zhou, L.L., Zhang, X., Guoming Xie, G.M., 2015a. Biosens. Bioelectron. 70, 351–357. Liu, R.J., Wang, Q., Li, Q., Yang, X.H., Wang, K.M., Nie, W.Y., 2017a. Biosens. Bioelectron. 87, 433–438. Liu, S.P., Su, W.Q., Li, Z.L., Ding, X.T., 2015b. Biosens. Bioelectron. 71, 57–61. Liu, W.P., Zhu, M.J., Liu, H.X., Wei, J.T., Zhou, X.M., Xing, D., 2016c. Biosens. Bioelectron. 81, 309–316. Liu, X., Shuai, H.L., Huang, K.J., 2015c. Anal. Methods 7, 8277–8284. Liu, X., Shuai, H.L., Liu, Y.J., Huang, K.J., 2016a. Sens. Actuators B-Chem. 235, 603–613. Liu, X., Zhang, J.Z., Huang, K.J., Hao, P., 2016b. Chem. Eng. J. 302, 437–445. Londin, E., Loher, P., Telonis, A.G., Quann, K., Clark, P., Jing, Y., Hatzimichael, E., Kirino, Y., Honda, S., 2015. Proc. Natl. Acad. Sci. USA 112, 1106–1115. Madalina, M.B., Christopher, M.A.B., 2016. Trends Anal. Chem. 79, 286–296. Mahin, K., Mohsen, B., Hossain, A.R.P., 2015. RSC Adv. 5, 35651–35660. Mariàngels, P.S., María, C.R., 2011. Anal. Chim. Acta 699, 134–152. Mas-Balleste, R., Gomez-Navarro, C., Gomez-Herrero, J., Zamora, F., 2011. Nanoscale 3, 20–30. Miao, P., Tang, Y.G., Zhang, Q., Bo, B., Wang, J., 2015a. ChemPlusChem 80, 1712–1715. Miao, P., Tang, Y.G., Yin, J., 2015b. Chem. Commun. 51, 15629–15632. Miao, P., Wang, B.D., Meng, F.Y., Yin, J., Tang, Y.G., 2015c. Bioconjugate Chem. 26, 602–607. Morrison, T.B., Weis, J.J., Wittwer, C.T., 1998. Biotechniques 24, 954–962. Nilsson, M., Gullberg, M., Dahl, F., Szuhal, K., Raap, A.K., 2002. Nucleic Acids Res. 30, 66. Paez, J.G., Lin, M., Beroukhim, R., Lee, J.C., Zhao, X.J., Richter, D.J., Gabriel, S., Herman, P., Sasaki, H., Altshuler, D., Li, C., Meyerson, M., Sellers, W.R., 2004. Nucleic Acids Res. 32, e7. Peeling, R., Mabey, D., 2010. Clin. Microbiol. Infect. 16, 1062–1069. Pei, H., Zuo, X.L., Pan, D., 2013. NPG Asia Mater. 5, 51. Peng, L.C., Zhang, P., Chai, Y.Q., Yuan, R., 2017. Anal. Chem. 89, 1338–1345. Peng, Y.L., Jiang, J.H., Yu, R.Q., 2014. Anal. Methods 6, 2889–2893. Popov, V., 2004. Mater. Sci. Eng. R.-Rep. 43, 61–102. Rajbir, S., Brinda, R., Suman, K., Arup, R.C., Saikh, J.H., Arnab, C., 2016. Cancer Lett. 38, 113–121. Richard, M.G., Ryan, C.B., 2016. Anal. Chem. 88, 431–450. Roya, T., Swahnnya, R.M.D., Almeida, J., Justin, G., 2015. Nanomed. Nanobiotechnol. 7, 580–592. Samanta, S., Balasubramanian, S., Rajasingh, S., Patel, U., Dhanasekaran, A., Dawn, B., Rajasingh, J., 2016. Trends Cardiovasc. Med. 26, 407–419. Shen, Y.T., Tian, F., ChenZ, Z., Li, R., Ge, Q.Y., Lu, Z.H., 2015. Biosens. Bioelectron. 71, 322–331. Shi, K., Dou, B.T., Yang, J.M., Yuan, R., Xiang, Y., 2017. Biosens. Bioelectron. 87, 495–500.

624