Magnetic nanoparticle-based amplification of microRNA detection in body fluids for early disease diagnosis

Abstract

Circulating biomarkers such as microRNAs (miRNAs), short noncoding RNA strands, represent prognostic and diagnostic indicators for a variety of physiological disorders making their detection and quantification an attractive approach for minimally invasive early disease diagnosis. However, highly sensitive and selective detection methods are required given the generally low abundance of miRNAs in body fluids together with the presence of large amounts of other potentially interfering biomolecules. Although a variety of miRNA isolation and detection methods have been established in clinics, they usually require trained personnel and often constitute labor-, time- and cost-intensive approaches. During the last years, nanoparticle-based biosensors have received increasing attention due to their superior detection efficiency even in very low concentration regimes. This is based on their unique physicochemical properties in combination with their high surface area that allows for the immobilization of multiple recognition sites resulting in fast and effective recognition of analytes. Among various materials, magnetic nanoparticles have been identified as useful tools for the separation, concentration, and detection of miRNAs. Here, we review state-of-the-art technology with regard to magnetic particle-based miRNA detection from body fluids, critically discussing challenges and future perspective of such biosensors while comparing their handling, sensitivity as well as selectivity against the established miRNA isolation and detection methods.

---

Isabel Gessner

Isabel Gessner studied chemistry at the University of Cologne and obtained her PhD degree in 2018 in Sanjay Mathur's group. Since then, she led a sub-group on nano–bio interactions under his supervision at the Institute of Inorganic and Materials Chemistry. She just recently transitioned as postdoctoral fellow to the Massachusetts General Hospital as part of the Harvard Medical School in Boston, USA. Her research focuses on the controlled synthesis of inorganic nanomaterials, especially magnetic nanoparticles, their surface modification and application in therapy, sensing, and imaging.

Jochen W. U. Fries

Jochen W. U. Fries is a Professor of Pathology at the University Hospital of Cologne with research interest in gene expression and microRNA regulation. He graduated in medicine from the Johannes Gutenberg University in Mainz and did most of his postdoctoral training and Pathology residency at the Harvard Medical School in Boston, USA from 1984 to 1997. His lab focusses on Endothelin-1 mediated signal transduction and regulation in chronic renal disease and renal tumors, characterizing new biomarkers, and their detection in urine and blood.

Veronika Brune

Veronika Brune obtained her PhD in chemistry in 2020 at the University of Cologne in the group of Sanjay Mathur. Her research focuses on the targeted synthesis and characterization of novel transition metal dichalcogenide (TMDC) single-source precursors with tunable properties and the employment of 2D van der Waals materials for renewable energy and bio-sensing applications.

Sanjay Mathur

Sanjay Mathur is a Chair Professor at the Institute of Inorganic Chemistry at the University of Cologne in Germany. He is a SPARC Faculty at the Indian Institute of Technology, Madras, India. He is also the Founding Director of the Institute of Renewable Energy Sources at the Xian Jiao Tong University, Xian, China and a World Class University Professor at the Chonbuk University in Korea. He is a Visiting Professor in the Institute of Global Innovation Research at TUAT, Japan. His research interests focus on application of nanomaterials and advanced ceramics for energy and biodiagnostic technologies. He holds ten patents and has authored/co-authored over 500 original research publications and has edited several books. He serves as the Editor for Nano Energy and Journal of Electroceramics. He is an Academician of the World Academy of Ceramics and Fellow of the American Ceramic Society and ASM International.

---

1. Introduction

A large portion of the human genome is considered to be potentially noncoding due to missing sequence homologies to known proteins, the absence of open reading frames, and frequent codon substitutions. They produce small, conserved ribonucleic acids, microRNAs (miRNAs), to control intracellular signaling,¹ and are involved in numerous mechanisms in health as well as disease.² MiRNAs are involved in development, cell differentiation, regulation of cell cycle,³ and consequently have been described in pathologic diseases of every major organ and condition e.g., cancer,^4–6 senescence,⁷ heart,⁸ neurodegenerative diseases,^9,10 gastrointestinal system,¹¹ and kidney/transplantation.¹²

MiRNAs have been used as diagnostic molecules in multiple studies, as noninvasive biomarkers either alone or in combination with other known disease predictors. However, as reviewed by Witwer,¹³ the reproducibility of biomarker studies is almost not existent, largely attributable to the non-standard terms of the studies, for instance, information on the procedure of detection is missing. Also, the standardization of assays in terms of detection sensitivity and specificity, expression levels, as well as compatibility between methods demands larger data sets as most of the published results involve small sample sizes. The availability of this information will then enable investigators to reliably compare profiles across various patient cohorts and demonstrate long-term outcome data and implications. Moreover, the detection of miRNAs from biological samples remains challenging due to their innately low concentrations and extensive interferences existing in body fluids. Nevertheless, given the possibility of magnetically concentrating samples, magnetic particles have demonstrated huge potential to overcome scalability issues with regard to extracting and quantifying disease-relevant, circulating biomarkers including antibodies,¹⁴ circulating tumor cells¹⁵ and nucleic acids.^16,17

This review highlights the difficulties arising from the diversity of methods currently used for miRNA isolation and detection. It presents the advantages in using surface-functionalized nanoparticles for enhancing the biorecognition event between miRNA and target ligands that results in an inherent local enrichment in point-of-care devices. Moreover, an easy separation of magnetic nanoparticles-miRNA conjugates by using an external magnetic field from serum, plasma or cell lysates enables superior detection capacity by reducing the interference with other biomolecules, such as proteins.¹⁸

2. MiRNAs

2.1. MiRNA biogenesis

MicroRNAs (miRNAs) are typically 19–22 nucleotide long, noncoding RNAs.^19,20 A primary miRNA (pri-miRNA) transcript is made through transcription by the RNA Polymerase II. By interaction with the so-called microprocessors, Drosha and DGCR8, a miRNA hairpin is generated, and exported to the cytoplasm via a complex of exportin-5 and GTP-bound Ran (RanGTP). After the nuclear export RNase III Dicer, in a complex with TAR RNA binding protein (TRBP) in a Dicer complex, processes this hairpin to result in a mature double-stranded miRNA. One strand in association with the Argonaute protein 2 forms the so-called miRNA induced silencing complex (miRISC). The remaining miRNA strand, miRNA*, also called passenger, is degraded. Within the RISC, the miRNA binds the target mRNA, based on the degree of complementarity. The complementarity of sequences between the miRNA and its target is preferentially located at the 5′ end of the miRNA, termed the seed, consisting of nucleotides 2–8, counting from the miRNA 5′ end.²¹ It leads to full mRNA degradation or blocking of the ribosomal machinery, both result in gene silencing.^22,23 MiRNA can bind at the 3′-untranslated region (3′-UTR), coding sequences (CDS) and 5′-untranslated region (5′-UTR), leading to the estimate of more than 60% of human mRNAs being regulated by miRNAs.^21,24 Binding of miRNA has been categorized into 5 classes, depending upon whether the seed sequence was used as well as on the complementarity of the nucleotides.²⁵

2.2. Body fluids as sources for miRNA detection

Recent studies have shown that “cell-free” miRNAs exist in all bodily fluids, which explains the growing interest in exploring their potential as disease markers.^26,27 Initially, miRNA expression was studied using different tissues (either fresh or paraffinized) or cell culture to determine functional and diagnostic roles of miRNAs.^28–30 However, bodily fluids are more readily available, less invasive than biopsies, and can be repeatedly taken for follow-up studies with reduced risks. Currently, body fluids from all different sources have been used.^31–36 Principally, they can be divided into 2 different types, depending on whether they can be obtained without invasive means (sputum, ejaculate, vaginal secret, menstruation blood, breast milk, colostrum, sweat, tears, urine, and feces) or only with invasive means (serum, plasma, cerebrospinal fluid, pleural fluid, peritoneal fluid, amniotic fluid, follicular fluid, synovial fluid, pancreatic juice, bile, gastric juice, and others) invasive means.^20,37 MiRNAs are secreted by cells through exosomes and extracellular vesicles.³⁸ It has been shown that the secreted miRNAs remain stable in bodily fluids.³⁹ Nevertheless, correlation of miRNAs in different body fluids under pathological conditions is still elusive and presumably not only dependent on the disease type but also on its stage (early or advanced) as well as patient-related aspects such as age, diet, medication, etc. For instance, in our previous study, Löser et al. demonstrated that levels of miRNA15a can significantly differ in blood/serum and urine. They showed that in proteinuric diseases high levels of miRNA15a could be detected in the urine, while the respective levels in serum were about 300 times lower.⁴⁰

2.3. Isolation methods

Isolation and accurate quantification of non-coding RNAs in body fluid samples is challenging due to the small effective concentration of small miRNAs and the coexistence of contaminating proteins and inhibitors, which may interfere with the isolation and detection of miRNAs.⁴¹ The most traditional approach is the use of a phenol-chloroform extraction followed by RNA precipitation.⁴² Lu and Rothenberg⁴³ described the loss of short structured RNAs (miRNAs, some pre-miRNAs, small interfering RNA (siRNA) duplexes, and transfer RNAs (tRNAs)) with low guanine–cytosine content (GC) due to the inefficiency of precipitating small versus long nucleic acid RNA as well as their lack of stability due to a stable secondary structure.⁴⁴ For these reasons, commercially available extraction kits^45–47 are used, which are more specialized in the miRNA extraction from different sample types. Even in currently available and applied bioanalytical procedures, lipoproteins and RNA–protein complexes may interfere with isolation, while co-isolation may falsify quantification affecting further applications.^48,49 Moreover, the hemolysis and cell remnants especially in blood samples may considerably change the miRNA yield.^50,51 In addition, available RNA quantification methods such as Qubit miRNA assay,⁵² the NanoDrop,⁵³ and the Bioanalyzer^54,55 vary in their quality of detection of low abundance targets, in comparative replicate extractions, and in the overall miRNA quantity measured.⁴¹

2.4. Current detection methods for miRNAs

Currently employed detection methods depend on the purpose of identifying miRNAs. All of these methods have in common that miRNA gene expression profiles are analyzed and compared between different disease and healthy states. Principally, one can differentiate between two types of detection approaches: (i) a screening assay to newly identify a broad range of simultaneously up-/down-regulated miRNAs (a miRNA microarray). These are done using either microarray assays or next generation sequencing. According to a very detailed study by Mestdagh et al., microarray assays often show low specificity, low concordance of differential expression, problems with sensitivity, specificity and compatibility, poor reproducibility, technology- vs. platform-related problems, and particularly exhibit a strong and significant inverse correlation between sensitivity and specificity.⁵⁶ Next generation sequencing is used as a screening tool because of its high throughput capabilities, however it needs highly specialized equipment to analyze complex data, which makes it more time consuming. Both methods need confirmation, for which quantitative real-time polymerase chain reaction (qRT-PCR) is chosen. (ii) The second type of detection focuses on individually expressed and regulated, already known miRNAs. There has been a multitude of techniques described as potential detection methods,³⁴e.g., the northern blot analysis with radiolabeled or biotinylated probes that involves size-selective differentiation to separate mature from pre-miRNA. The in situ hybridization is suitable for highly abundant miRNAs and is generally inconsistent for miRNAs in low abundance. Other alternative techniques involve bioluminescence miRNA detection as competitive solid-phase hybridization-based method,⁵⁷ surface-enhanced Raman spectroscopy method,⁵⁸ surface plasmon resonance imaging (SPRI)-based,⁵⁹ and quantitative real-time PCR (qRT-PCR).^60,61 For a detailed comparison of current miRNA detection methods with other parallel approaches, including comprehensive overviews on their advantages and disadvantages, we refer to other specialized reports and reviews.^62–64

Over the last 10 years, the most commonly used assay technique has been the qRT-PCR analysis. According to Lu and Rothenberg,⁴³ three different approaches can be chosen:

(i) Reverse transcription with miRNA specific stem-loop primer, followed by qPCR using a miRNA specific probe. This is the more expensive variant, but it can differentiate between pri-, pre- and mature miRNA. In addition, stem loop primers prevent the biding of double-stranded genomic DNA molecules as well as enhancing the thermal stability of RNA–DNA heteroduplexes.⁶⁵ If the material quantity is small, pooled stem-loop primers may provide advantages,⁶⁶ while pre-amplification may help if accuracy is the primary question.⁶⁷

(ii) Universal reverse transcription by adding polyA tails to 3′ end of miRNA, followed by reverse transcription with a poly T primer with a universal sequence (Tag) appended at the 3′ end.

(iii) Universal reverse transcription by adding a 5′ adapter sequence and polyA tails to 3′ end of miRNA, followed by reverse transcription with a poly T primer and a universal sequence (Tag) appended at the 3′ end. Adding a polyA tail is cost-efficient, and useful if different miRNAs have to be analyzed in a small amount of starting material but it does not allow differentiation between precursor miRNAs.

The current setbacks of this method are at least threefold: (i) extraction of miRNA is necessary (high vs. low abundance being an issue); (ii) varying composition in the SYBR Green solution; and (iii) the detection device for qRT-PCR itself. There is currently no comparative analysis available for any of these three potential problems, so that qRT-PCR may yield different mRNA levels even if the identical material source is used. According to the current state-of-the-art, such variations cannot be tolerated in a reliable and meaningful diagnostic test.

In a recent review, Fauth et al.⁶⁸ have outlined basic criteria for the validation of miRNA quantification assay using qRT-PCR. Several criteria have to be met as outlined below:

(1) Linearity of quantification within the calibration range of measurement, including the smallest accurately measurable concentration standard (lower limit of quantification, LLOQ), being measured on three different days.⁶⁹

(2) Sensitivity by determining the LLOQ.

(3) Recovery: samples being spiked with known miRNA concentrations before and after RNA isolation followed by qRT-PCR.⁷⁰

(4) Precision being defined as repeatability (intraday-precision) and reproducibility (interday-precision) by three independent runs.⁷¹ According to the literature, variations of 25% for intraday-precision and 35% for interday-precision are acceptable.^69,70

(5) Accuracy means the deviation of the measured concentration from the nominal concentration, ±25% being acceptable for the detection of nucleic acids.⁷²

(6) Matrix effect was determined by spiking water and plasma with miRNA at three different concentrations, where quantification cycle (C_q)-values of water were compared with those in the matrix.

(7) Stability was analyzed in plasma versus RNA-isolate in DNAse-free water stored up to one month (−80 °C), and complementary DNA (cDNA) at −20 °C for one month.

Normalization with reference genes⁷³ has to regard the tissue specificity of miRNAs from where it originates, which should be assessed in the tissue of interest before start of the analyses.⁷⁴ Selection of the most stably expressed miRNAs seem to provide the best non-functional reference genes independent of disease stage.⁷⁵ Run-to-run variation is composed of instrument related variation (PCR block, lamp, filters, etc.), data analysis settings (baseline correction and threshold), reagents (polymerase, fluorophores, etc.) and the optical properties of plastics. The amount of input RNA required varies substantially between platforms and could be a limiting factor when choosing a platform. Ultimately, the RNA quantity and quality may dictate the platform used to obtain the extracellular miRNA profile. In the end, detection rate is a combination of sensitivity, detection cutoff, and platform content (number of assays or probes available on the platform).

3. Nanomaterial-based miRNA biosensors

The possibility of selectively sensing and analyzing single molecules or cells provides rich information on the quantity and presence of rare and aberrant species that would otherwise remain unseen given the complex composition of biological samples and the co-existence of several biomolecules.⁷⁶ MiRNAs are intrinsically involved in all biological processes and omnipresent in body fluids, which makes their selective capture and detection critically important for enhancing the diagnostic and prognostic capabilities of bio-analytical assays. However, the visualization of potential heterogeneities in biological species, including miRNAs, are of critical importance for the diagnosis and progression monitoring of diseases at very early stages. Therefore, while much progress has been made in the development of biosensing platforms capable of delivering a signal specific to only one type of species, other analyte signals remain nonspecific and thus do not interfere with the bioassays. In the presence of the target analyte, a biorecognition event takes place that is transferred to a transducer for a quantifiable signal.⁷⁷ The choice of the biorecognition element and the transducer is, hence, crucial to obtaining biosensors that possess high selectivity, sensitivity, reproducibility and the reusability needed for potential point-of-care analysis of biomarkers.

Nanomaterials possess promising potential for the detection of analytes due to their intrinsic electrical, magnetic, or optical properties and, more importantly, due to their high surface area that can be exploited for the immobilization of target ligands with high sensitivity towards target biomolecules.⁷⁸ Multiple approaches for the sensing of oligonucleotides have been reported based on metal or metal oxide nanostructured films.^79,80 Compared to thin films or particles immobilized on a solid surface, where diffusion of analytes and contact to recognition sites on the sensor surface can be very tedious, dispersible sensor particles can provide much faster detection times and higher efficiencies due to enhanced probability of ligand-receptor collisions. In fact, biosensors based on dispersible nanoparticles have shown 1000 times superior response time and detection limit.⁷⁶ The latter aspect can be attributed to the availability of the large number of recognition sites on the nanoparticle surface, which results in higher sensitivities due to increased density of biomolecule capturing events.

Nanoparticles generally demonstrate ease of functionalization though the abundance of surface-active chemical groups, e.g., surface hydroxyl groups in terms of metal oxide nanoparticles. Immobilization of target-specific ligands to nanoparticles through surface groups can occur either covalently or non-covalently, e.g., through electrostatic interactions, π–π stackings or van der Waals forces.⁸¹ Although unspecific physisorption can result in higher numbers of attached ligands, the number and orientation of surface-bound molecules can be precisely controlled during chemical conjugation reactions.⁸² Most cases of miRNA sensors rely on the attachment of single-strand DNA or RNA, complementary to the analyte sequence, to nanoparticle surfaces. Conjugation strategies therefore often involve carbodiimide or click reactions. Nevertheless, most approaches especially those based on magnetic nanoparticles involve the immobilization of streptavidin, which demonstrates an exceptional high binding affinity (dissociation constant K_d ∼ 10⁻¹⁵ M) to biotin, which thus represents one of the strongest non-covalent interactions known in nature.⁸³ The selectivity of miRNA biosensors is usually tested by employing single-nucleotide mismatches.

Much effort has been devoted to the controlled engineering of materials at the nanoscale to obtain efficient miRNA sensor platforms that could be used in future point-of-care diagnostic devices. As a result, various nanomaterials including magnetic particles and composites thereof have been incorporated into sensing applications and devices used for the detection of miRNAs (Fig. 1), providing enhanced sensitivities and reduced detection times compared to traditional detection kits.^84,85

Fig. 1 Overview of functional properties of magnetic nanoparticles and their composites with other nanomaterials as used for miRNA sensing applications.

3.1. Magnetic nanomaterials in miRNA sensors

Given the generally low concentration of miRNAs in body fluids, potential strategies for the enrichment of miRNAs in the samples, e.g., through use of magnetic extraction represents a powerful tool for the detection of biomarkers and is of significant importance for reliable and early diagnosis of various diseases.⁸⁵ Besides the material composition, the employed particle size is of critical importance in the context of their efficient employment in extraction and biosensing devices. In fact, most magnetic separation studies rely on (sub)micron particles used for their high magnetic moment that allows for rapid extraction of analytes.⁸⁶ However, besides being less prone to sedimentation, smaller (<100 nm) nanoparticles offer a higher surface area that enables attachment of large amounts of analyte recognition sites leading to high sensitivities. Nevertheless, if employed particles are below a certain critical size range (usually below 20 nm), the viscous drag, predominantly present in body fluids, dominates the magnetophoretic force and hampers their diffusion, which leads to unwanted prolongation of the separation process.⁸⁷ The use of magnetic particles in miRNA sensors therefore also provides the possibility to concentrate larger sample volumes, which is beneficial compared to commonly used small volumes in the μl range as it allows for a simple handling and is much more representative of the bulk population.⁷⁶

During the past decade, magnetic nanoparticles have been efficiently used in numerous miRNA sensors. In many cases, optical or electrochemical readout techniques are employed providing ease of handling. Therefore, combining the beneficial properties of two or more material types has been proven very advantageous for biosensing applications.

3.2. Plasmonic noble metal nanoparticles, quantum dots and their composites with magnetic particles in miRNA sensors

Noble metal nanoparticles such as gold and silver nanoparticles have been exhaustively used for the sensing of various biological analytes mainly due to their size- and shape-dependent surface plasmon resonance (SPR), the collective oscillations of metal conduction band electrons with visible light photons.^88,89 Gold nanoparticles represent a highly versatile tool in biosensors for the detection of numerous disease-related miRNAs¹⁸ from body fluids including blood,⁹⁰ urine,⁹¹ and saliva.⁹² Besides simple metal nanoparticles, composite structures have demonstrated high efficiency in miRNA detection, e.g., using Ag/Pt nanoclusters for miRNA-21 sensing in human urine samples.⁹³ A recently developed biosensor for miRNA-133a, an acute myocardial infarction-related miRNA, employed hollow Au/Ag nanospheres using surface-enhanced Raman scattering (SERS) as a highly sensitive detection method.⁹⁴ Moreover, during the last few years fluorescent nanobiosensors for miRNAs have been developed based on the tunable optical properties of quantum dots, e.g., made of carbon⁹⁵ or CdTe.⁹⁶

A broad variety of composite structures with magnetic nanoparticles has been reported for miRNA sensing, e.g., using assemblies of nano-γ-Fe₂O₃ particles in combination with Pt nanoparticles for the detection of miRNA-21,⁹⁷ polydopamine-coated Fe₃O₄ nanoparticles and carbon dots for sensing of miRNA-167,⁹⁸ or carboxyl-terminated Fe₃O₄ nanoparticles in microfluidic concentrator devices to quantify miRNA-200a-3p with silica embedded Au nanoparticles that function as Raman tags.⁹⁹ Moreover, Fe₃O₄@Ag core–shell nanoparticles have been employed for miRNA let-7b capture and quantification with sensitivities in the lower fM range using SERS as a detection method.¹⁰⁰ Even higher sensitivities with detection limits of miRNAs in the aM range could be recently obtained using Fe₃O₄@Au nanomaterials employing a DNA four-way junction-based electrochemical detection of target miRNA.¹⁰¹ The selectivity of the used system was employed using six different mismatched miRNAs and demonstrated similar results to qRT-PCR on healthy and breast cancer patient samples (Fig. 3).

Moreover, a multicomponent sensor consisting of magnetic nanoparticles for magnetic extraction of analytes, silver nanoparticles for chemical amplification and CdTe quantum dots as luminescent signals were reported for the detection of miR-141 in the femtomolar concentration range.¹⁰²

3.3. Two-dimensional nanostructures and their composites with magnetic particles in miRNA sensors

Few layer graphene oxide (GO) that characterizes highly dispersed and stable two-dimensional (2D) carbon material, represents one of the most commonly employed 2D structures in medicinal applications including miRNA biosensors, based on its unique optical properties, extremely high specific surface area and π-conjugated structure.^103,104 Hizir and co-workers demonstrated the versatility of graphene oxide for miRNA sensing applications and reported on the efficient simultaneous detection of exogenous miRNA-21 and miRNA-141 from human body fluids including blood, urine, and saliva.¹⁰⁵ Additionally, two-dimensional transition metal dichalcogenides (TMDCs; e.g., MoS₂, WS₂, etc.) have shown great potential for the sensing of miRNAs owing to the high affinity of sulfur that allows simple surface functionalization, e.g., through non-covalent self-assembly of thiolated compounds including miRNAs on its surfaces (Fig. 2). For instance, Xi and co-workers employed WS₂ nanosheets for the detection of miRNA-21 through fluorescence quenching,¹⁰⁶ while Cai et al. reported on the suitability of MoS₂ for sensing miRNA-21 from serum samples.¹⁰⁷

Fig. 2 SEM images of (A) Fe₃O₄ and (B) Fe₃O₄@Au. (C) Comparison of peak current variations for the detection of mismatch miRNAs with and without spiking target miRNA. (D) Quantification of miRNA concentrations in serum samples from (a and b) breast cancer patients and (c and d) healthy individuals. Reprinted from Yu et al.¹⁰¹ Copyright (2020) with permission from Elsevier.

Fig. 3 Fluorescence quenching capabilities of two-dimensional transition metal dichalcogenides observed upon adsorption of dye-labeled RNA has been commonly exploited for miRNA sensors, since duplex formation with target miRNAs leads to desorption and restorage of the fluorescent signal.

Magnetic composites with 2D materials have been mainly reported for graphene-based miRNA sensors, although studies employing TMDCs have been described, even in body fluids (see Section 4.2). For instance, GO-loaded superparamagnetic iron oxide nanoparticles have been employed for the electrocatalytic detection of cancer-related miRNA-21 with low detection limit of 1.0 fM.¹⁰⁸ Another approach used GO-magnetic microparticle hybrids employing chemiluminescence as detection method for miRNA-122 sensing in the lower picomolar range.¹⁰⁹ A complex sensor design employing reduced GO as sensing platform in combination with Au nanoparticles and CdSe@CdS quantum dots-contained magnetic nanocomposites was presented by Daneshpour and coworkers for dual-signal simultaneous detection of miRNA-106a and let-7a.¹¹⁰

4. Magnetic particle-based miRNA isolation and detection from body fluids

Even though the efficiency of magnetic nanoparticles and its composite structures for miRNA sensing applications has been demonstrated, most experiments were performed in the absence of potentially interfering biomolecules, and thus only few studies have been performed in natural body fluids. Still, the fast, sensitive, and selective extraction of analytes in biologically complex media is the key challenge to receive highly efficient biosensing platforms that enable their simple handling and short processing times.

Here, we highlight some of the recent approaches that used magnetic particles for the detection, separation and quantification of circulating miRNAs out of body fluids including serum, whole blood and urine (Table 1). The chosen sensing platforms all incorporate magnetic structures or their composites in the nano to lower micron range.

Table 1 Overview of recently demonstrated magnetic particle-based approaches for the detection of miRNAs from body fluids

Employed material	Target miRNA(s)	Tested body fluid(s)	Signal readout	Limit of detection	Linear range	Ref.
Streptavidin-coated magnetic microparticles	miRNA-141	Serum	Fluorescence	46 pM	50 pM–5 nM	111
Fe2O3@Au composite nanoparticles	miRNA-21	Serum	Colorimetric	1 aM	1 aM–5 pM	112
Streptavidin-coated magnetic microparticles	miRNA-21	Serum	Mass spectrometry (LC-ESI-MS/MS)	60 fM	0.2 pM–0.25 nM	113
Fe3O4@SiO2 microsphere–rGO composite	miRNA-21	Serum	Fluorescence	0.098 nM	0.2–20 nM	116
Two-dimensional Fe3O4/DNA network structures	miRNA let-7a	Serum	Colorimetric	13 aM	0.05 fM–12 nM	117
Multilayer core–satellite magnetic superstructures	miRNA let-7b	Serum	Optomagnetic	4.8 fM	10 fM–10 nM	118
Fe3O4@Au composite nanoparticles	miRNA-21 and miRNA-155	Serum	Electrochemical	1.5 fM; 1.8 fM	5 fM–2 nM	120
Streptavidin-coated magnetic microparticles	miRNA-122, miRNA-155 and miRNA-21	Serum	HPLC-fluorescence	0.39 fM; 0.30 fM; 0.26 fM	1 fM–100 pM	121
Quantum dot decorated magnetic beads	miRNA-16 and miRNA-296	Serum	Colorimetric	0.5 pM	0.5 pM to 10 pM	122
Fe3O4@Au composite nanoparticles	miRNA-21	Whole blood	Electrochemical	10 aM	10 aM–1 nM	123
Magnetic beads coupled with gold nanoparticles	miRNA-141	Whole blood	Photoelectrochemical	0.5 fM	1 fM–10 pM	124
MoS2 sheets modified with CuFe2O4 and streptavidin-coated magnetic beads	miRNA-205, miRNA-92, miRNA-7857 and miRNA-378	Blood	Electrochemical	0.48 pM	1 pM–1.5 nM	125
Streptavidin-coated magnetic beads	miRNA-21	Urine	Enzymatic conversion of sucrose to glucose	1.8 pM	10–200 pM	130
Carboxylic acid modified magnetic microparticles	miRNA-21	Urine	Magnetic relaxation switch	5 fM	5 fM–0.5 nM	131
Fe3O4 nanoparticles in combination with a capped gold nanoslit	miRNA-16-5p	Urine	SPR	17 fM	1 pM–1 nM	132

---

4.1. Detection from serum

The detection of small amounts of miRNAs from serum samples usually requires amplification of analytes and corresponding signals, similar to qRT-PCR, which is the current gold standard, for miRNA expression profiling analysis. The required concentrations in a typical serum analysis usually range between 10⁻¹⁶–10⁻¹² M but can be even lower.⁷⁶ Besides using magnetic concentration as an amplification approach, proportional and exponential amplification strategies based on circulatory systems have been developed in miRNA biosensing devices. DNA-based cycles usually rely on the modification of nanoparticles with specific DNA sequences that are displaced in parts as soon as the analyte is present and attached. Released molecules are then detected, e.g., based on fluorescence measurements as described for the detection of miRNA-141 out of serum samples using DNA modified magnetic beads.¹¹¹ Here, binding of the target DNA to modified magnetic beads resulted in a toehold-mediated DNA displacement reaction, creating an intermediate toehold structure. The addition of a fuel DNA then triggered the release of miRNA-141 and a dye labeled DNA, which could be fluorescently detected while the target miRNA could undergo the same proportional DNA cycle leading to signal amplification (Fig. 4). If a trigger DNA with the identical sequence to miRNA-141 is additionally employed, signal and target amplification can be exponential. Direct comparison between proportional and exponential amplification revealed a higher sensitivity and shorter processing times in terms of the exponential amplification, although a higher noise fluorescence signal was observed.

Fig. 4 Schematic illustration of the principles of DNA circuit systems based on (a) proportional and (b) exponential amplification strategies. Reprinted from ref. 111 (https://pubs.acs.org/doi/full/10.1021/acsomega.7b01866) with permission from ACS. Further permissions related to the material excerpted should be directed to the ACS.

A few exponential amplification strategies were also employed on composite structures using gold decorated Fe₂O₃ nanoparticles as detection platforms.¹¹² Using this approach, attomolar detection limits of miRNA-21 in serum samples were achieved based on a highly sensitive calorimetric approach. Alternatively, target amplification can be performed enzymatically through duplex specific nuclease (DSN) addition. DSN demonstrates strong cleavage preference for DNA in double-stranded nucleic acids (e.g., DNA/RNA heteroduplexes), while being able to differentiate perfectly matching duplexes from imperfect matches. Therefore, DSN will selectively cleave duplexes formed upon hybridization of target miRNA with DNA modified magnetic beads. The released target miRNA can be recycled and undergo the same process several times to amplify the analyte signal. This approach was recently demonstrated for the quantification of miRNA-21 out of serum samples using liquid chromatography-electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) and DNA modified magnetic beads as platform.¹¹³ Magnetic separation of DNA fragments that were formed through the addition of DSN enabled their concentration followed by acid hydrolyzation into nucleobases before LC-ESI-MS/MS quantification. The authors chose miRNA-21 as target analyte which represents one of the earliest identified cancer-promoting ‘oncomiRs’, oncogenic miRNAs that are usually overexpressed in cancers,¹¹⁴ is related with multiple cancer types, and thus constitutes an important diagnostic and prognostic biomarker.¹¹⁵ Sensing of this target was also demonstrated employing a composite structure of silica (SiO₂) coated magnetic Fe₃O₄ particles and reduced graphene oxide (rGO) using the fluorescence quenching ability of rGO as read-out.¹¹⁶ Although high selectivity compared to other miRNAs could be demonstrated, obtained detection limits remained in the lower nM range.

Attomolar concentrations of miRNA let-7a, which is part of the lethal-7 (let-7) family and is suppressed in tissues of gastric cancer patients, could be detected using two-dimensional DNA/Fe₃O₄ networks displaying triple amplification via synergy of the amplification events hybridization chain reaction, networking and catalytic reaction.¹¹⁷ Validation of this sensor was additionally approved in serum samples of gastric cancer patients and healthy people demonstrating its practicability in real body samples. An optomagnetic readout for sensing let-7b was described using ‘core–satellite’ magnetic superstructures that comprise multiple layers of differently sized magnetic beads.¹¹⁸ The presence of the target miRNA triggers release of satellite magnetic particles though addition of DSN, which can be quantified using an optomagnetic signal. This simple approach provided high sensitivities in the lower femtomolar range and could be used in serum samples. Chan and coworkers just recently presented a novel biosensor based on silica coated magnetite nanoparticles employing a ligase-mediated amplification for the detection of miRNA-149 in human serum samples.¹¹⁹ Since miRNA-149 has been reported to be involved in multiple cancer types, the proposed sensor was evaluated in both healthy and cancer patient's serums reliably quantifying circulating miRNA-149 with a detection limit of 314 fM.

While the above-mentioned approaches demonstrate the feasibility of single miRNA detection with high sensitivity and selectivity in serum samples, the coincident detection of multiple target analytes still represents a major challenge, also considering particle pre-modification. Nevertheless, Shen and co-workers just recently presented an electrochemical biosensor based on magnetic Fe₃O₄@Au composite nanoparticles capable of simultaneously detecting miRNA-21 and miRNA-155, which are both overexpressed in the serum of patients with breast cancer.¹²⁰ For this purpose, electrochemical tag-labeled DNAs were immobilized on the surface of nanoparticles via a hyperbranched hybridization chain reaction. Upon attachment of target analytes, both miRNAs were magnetically concentrated and quantified up to detection limits of 1.5 fM (miRNA-21) and 1.8 fM (miRNA-155), respectively, in serum samples. By use of high-performance liquid chromatography (HPLC)-fluorescence as detection method, Qi and coworkers, reported on the simultaneous detection of miRNA-122, miRNA-155, and miRNA-21 in serum samples from healthy persons and cervical cancer patients.¹²¹ High sensitivities in the lower fM range were obtained using a DSN-assisted target recycling amplification. Therefore, magnetic beads were functionalized with oligonucleotide sequences complementary to target miRNAs and upon hybridization, DSN was added to selectively cleave heteroduplexes and release the target miRNAs which were separated and quantified by HPLC-fluorescence. Their results indicated that miRNA-155 and miRNA-21 were significantly overexpressed in cancer patients compared to those of healthy persons, which was consistent with qRT-PCR amplification experiments. Moreover, Wang and coworkers recently presented the employment of magnetic beads that were surface decorated with quantum dots for the simultaneous detection of miR-296 and miR-16. Detection of both miRNAs with femtomolar sensitivity and single-base mismatch specificity in cells and in diluted serum samples provides a novel approach for early stage diagnosis of hand, foot, and mouth disease.¹²²

4.2. Detection from whole blood

Tavallaie et al. were the first to describe a nanoparticle-based biosensor that was able to detect miRNAs in a range of 10 aM to 1 nM in unprocessed blood samples.¹²³ Detection of target miRNA was performed using gold-coated magnetic nanoparticles (Au@MNPs) as platform by surface modifying them via an Au–S bond with a short DNA sequence complementary to the sequence of target miRNA. Moreover, a methylene blue redox label at the end of the DNA served as a detecting agent upon miRNA–DNA hybridization and subsequent magnetic concentration of conjugates on the surface of a gold microelectrode measuring square-wave voltammetry.

Based on the fact that circulating miRNA-141 in blood samples of patients has been recognized as potential biomarkers for prostate carcinoma, Zhang and co-workers developed a biosensor based on magnetic beads and gold nanoparticles capable of quantifying the target miRNA in unprocessed blood samples.¹²⁴ In the presence of the target molecule, selective binding to magnetic beads that were surface decorated with DNA modified gold nanoparticles occurred. Magnetic separation was carried out followed by the addition of a sufficient fuel strand that led to the release of miRNA-141 as well as DNA-modified gold nanoparticles. The latter bound with complementary sequences on a photoactive polymer dots/indium tin oxide (ITO) electrode allowing to measure a decrease in photocurrent caused by the energy transfer between gold nanoparticles and polymer beads which directly correlated to the amount of target miRNA present in tested blood samples.

Chand and co-workers reported the use of MoS₂ nanosheets modified with CuFe₂O₄ nanoparticles, which functioned as electrochemical amplifiers, in combination with streptavidin functionalized magnetic beads for miRNA detection from body fluids.¹²⁵ The sensor was designed to detect paratuberculosis, a bacterial infection of the intestinal tract of dairy cattle, by detecting four different miRNAs (miRNA-205, miRNA-92, miRNA-7857, and miRNA-378) simultaneously. While a detection limit of 0.48 pM could be obtained in spiked serum samples, their functionality was additionally proven in bovine blood samples although not a complete, albeit good, agreement with ELISA and qRT-PCR results could be obtained.

4.3. Detection from urine

Besides the fact that a variety of studies have demonstrated that miRNAs released via urine can provide a wealth of information useful to diagnose multiple diseases like urinary system cancer, kidney injury, and diabetes,^126–128 only few miRNA sensing systems based on magnetic nanomaterials have been developed so far to extract and detect miRNAs from urine samples. Although commercial kits such as urine microRNA purification kit (Norgen Biotek) or miRNeasy kit (Qiagen) exist, Xu and co-workers recently demonstrated that the use of magnetic nanoparticles led to 25–30% improved miRNA yields besides saving reagents and time.¹²⁹ Three types of urinary disease biomarkers, namely miRNA-10a, miRNA-16, and miRNA-30d, were extracted from rat urine upon enrichment using carboxyl-coated magnetic nanoparticles. Since most miRNAs in urine remain protected from RNAses either enclosed in microvesicles such as exosomes or complexed in Argonaute 2 (Ago2)-containing protein complexes, the enrichment was based on a protein corona formation on the surface of magnetic nanoparticles. A more specific miRNA detection approach from urine samples was demonstrated by Huang and co-workers, who developed biosensors based on magnetic nanoparticles for the detection of miRNA-21 from mouse urine using glucose meters.¹³⁰ In this study, streptavidin coated magnetic nanoparticles were functionalized with a biotinylated DNA strand to which invertase was linked. Upon binding of target miRNA-21 via DNA/RNA hybridization, DSN was applied, resulting in the release of invertase which efficiently converts sucrose to glucose. After magnetic separation, and addition of sucrose to the supernatant, measured glucose levels thus directly correlated with miRNA-21 levels with a detection limit of 1.8 pM. Besides the high sensitivity and high sequence selectivity, this approach demonstrates a convenient and low-cost strategy using a simple glucose meter as signal readout device. The magnetic properties of employed nanoparticles are not only very beneficial for analyte enrichment purposes but can also be used for the detection itself. In this regard, Lu et al. recently employed magnetic microparticles, combining magnetic separation with visualization of magnetic relaxation changes that enabled rapid and efficient sensing of miRNA-21 in urine samples.¹³¹ The significant change of transverse relaxation time (ΔT₂) that was observed upon binding of the target miRNA to the particle surface functioned as signal readout to quantify the oligonucleotide at concentrations of 5 fM.

A microfluidic system comprised of magnetic nanoparticles and gold surfaces which enabled magnetic separation followed by detection of analytes via surface plasmon resonance (SPR) was demonstrated by Mousavi et al. who focused on the detection and quantification of miRNA-16-5p, a prognostic indicator for acute kidney injury, in human urine samples (Fig. 5).¹³² For this purpose, magnetite nanoparticles were surface modified with an oligonucleotide sequence complementary to a unique region of the target miRNA-16-5p sequence via a sulfhydryl-reactive cross-linker. Upon capturing of the target molecule, hybridization occurred with a second complementary strand that was linked to a gold surface and allowed for the visualization and quantification of attached miRNA by SPR.

Fig. 5 A schematic of the double hybridization method to detect the nucleic acid target. (a) First step: isolation of the target molecule using magnetic nanoparticles. (b) Second step: hybridizing the target molecule isolated on the magnetic nanoparticles with probe II on the capped gold nanoslit. Republished from Mousavi et al.¹³² with permission from The Royal Society of Chemistry, Copyright (2015).

5. Conclusion and future perspectives

The detection and quantification of specific miRNAs as circulating biomarkers from body fluids carries enormous potential as diagnostic biomarkers for a variety of disorders including the early diagnosis life-threatening disease such as cancers and/or ailments with hardly predictable course. However, their small size, high sequence homology among family members, and low abundance in body fluids bedevils their detection in short time and with high accuracies that poses a challenge on miRNA-based bioanalytical approaches. In order to overcome cumbersome and tedious steps of traditional extraction processes, biosensors based on nanomaterials have been developed that offer enhanced miRNA yields and reduced costs. Especially magnetic materials have received increasing attention due to the possibility of concentrating the analyte through specific surface conjugation that has served as the basis for a number of highly efficient miRNA sensing devices. Nevertheless, the current challenge lies in the demonstration of the miRNA capture and detection strategies in real biological samples, which is a prerequisite for the development of point-of-care testing systems. While most sensors demonstrate high sensitivities and selectivity in simple water-based model assays, the presence of other interfering biomolecules in conjunction with the low number of testing analyte, increasing viscosity of the sample and possible influence due to pH changes might affect many detecting systems. Some of the approaches highlighted in this work display the efforts that have been made in capturing and quantifying biomarker miRNAs from (unprocessed) biologically complex body fluids with very low detection limits down to fM and even aM levels and fast processing times, e.g., using electrochemical or magnetic relaxation readouts. In view of the biochemical complexity of many diseases, simultaneous detection of multiple miRNAs, potentially even obtained from different body fluids, would be of high interest to enable an accurate disease diagnosis. In this context, since current researches focuses on the correlation of miRNA expression in different body fluids, it might be of particular interest to develop magnetic amplification-based miRNA sensors that go beyond the tested fluids such as serum, blood, and urine.

Although some sensors still rely on specific analytical equipment that requires trained personnel, first integrations into low-cost signal readout devices, such as glucose meters demonstrate a convenient alternative for patient-centric point-of-care devices. These methods now have to be optimized towards their collection, processing efficiency and reproducibility to be suitable for reliable disease diagnosis (Fig. 6).

Fig. 6 Future steps on the way to efficient and portable miRNA sensing devices based on magnetic nanoparticles.

Moreover, while some miRNAs are free-floating in biological fluids, many are contained in extracellular vesicles, such as exosomes. Therefore, current research points towards the development of biosensors capable of capturing those microvesicles and detecting the encapsulated oligonucleotide which is expected to provide new and highly promising opportunities regarding disease diagnosis.^133,134

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Authors are thankful to the University of Cologne for the infrastructural and financial support. Dr Isabel Gessner is thankful to the German Science Foundation (DFG) for the award of a Walter Benjamin Fellowship (2020#x2013;21) to her.

References

1. R. P. Alexander, G. Fang, J. Rozowsky, M. Snyder and M. B. Gerstein, Nat. Rev. Genet., 2010, 11, 559–571.
2. D. Trzybulska, E. Vergadi and C. Tsatsanis, eJIFCC, 2018, 29, 221–226.
3. D. P. Bartel, Cell, 2009, 136, 215–233.
4. Y. Tutar, Curr. Pharm. Biotechnol., 2014, 15, 429.
5. M. Cui, H. Wang, X. Yao, D. Zhang, Y. Xie, R. Cui and X. Zhang, Front. Genet., 2019, 10, 626.
6. X. Zhou, Z. Zhang and X. Liang, Cell J., 2020, 21, 459–466.
7. R. Munk, A. C. Panda, I. Grammatikakis, M. Gorospe and K. Abdelmohsen, Int. Rev. Cell Mol. Biol., 2017, 334, 177–205.
8. A. Wojciechowska, A. Braniewska and K. Kozar-Kamińska, Adv. Clin. Exp. Med., 2017, 26, 865–874.
9. M. Ferrante and G. O. Conti, MicroRNA, 2017, 6, 157–165.
10. Z.-G. Liu, Y. Li, J.-H. Jiao, H. Long, Z.-Y. Xin and X.-Y. Yang, Neural Regener. Res., 2020, 15, 2123–2130.
11. R. Nedaeinia, M. Manian, M. H. Jazayeri, M. Ranjbar, R. Salehi, M. Sharifi, F. Mohaghegh, M. Goli, S. H. Jahednia, A. Avan and M. Ghayour-Mobarhan, Cancer Gene Ther., 2017, 24, 48–56.
12. K. J. Ledeganck, E. M. Gielis, D. Abramowicz, P. Stenvinkel, P. G. Shiels and A. H. Van Craenenbroeck, Clin. J. Am. Soc. Nephrol., 2019, 14, 454–468.
13. K. W. Witwer, Clin. Chem., 2015, 61, 56–63.
14. S. Yadav, M. K. Masud, M. N. Islam, V. Gopalan, A. K. Y. Lam, S. Tanaka, N. T. Nguyen, M. S. Al Hossain, C. Li, M. Y. Yamauchi and M. J. A. Shiddiky, Nanoscale, 2017, 9, 8805–8814.
15. S. Bhana, Y. Wang and X. Huang, Nanomedicine, 2015, 10, 1973–1990.
16. N. Ali, R. D. C. P. Rampazzo, A. D. T. Costa and M. A. Krieger, BioMed Res. Int., 2017, 2017, 9306564.
17. I. Gessner, X. Yu, C. Jüngst, A. Klimpel, L. Wang, T. Fischer, I. Neundorf, A. C. Schauss, M. Odenthal and S. Mathur, Sci. Rep., 2019, 9, 1–10.
18. C. Coutinho and Á. Somoza, Anal. Bioanal. Chem., 2019, 411, 1807–1824.
19. O. J. Snove and J. J. Rossi, Genome Biol., 2006, 7, 231.
20. M. von Brandenstein, C. Richter and J. W. U. Fries, Life Sci., 2012, 91, 475–489.
21. T. M. Witkos, E. Koscianska and W. J. Krzyzosiak, Curr. Mol. Med., 2011, 11, 93–109.
22. K. Sun and E. C. Lai, Nat. Rev. Genet., 2013, 14, 535–548.
23. A. Vishnoi and S. Rani, Methods Mol. Biol., 2017, 1509, 1–10.
24. P. Maziere and A. J. Enright, Drug Discovery Today, 2007, 12, 452–458.
25. A. M. Mohr and J. L. Mott, Semin. Liver Dis., 2015, 35, 3–11.
26. S. Junqueira-Neto, I. A. Batista, J. L. Costa and S. A. Melo, Acta Cytol., 2019, 63, 479–488.
27. S. Kumar, M. Vijayan, J. S. Bhatti and P. H. Reddy, Prog. Mol. Biol. Transl. Sci., 2017, 146, 47–94.
28. E. Azzalini, E. De Martino, P. Fattorini, V. Canzonieri, G. Stanta and S. Bonin, Int. J. Mol. Sci., 2019, 20, 4819.
29. M. I. K. Anindo and A. Yaqinuddin, Int. J. Surg., 2012, 10, 443–449.
30. A. M. L. Coenen-Stass, M. J. Pauwels, B. Hanson, C. Martin Perez, M. Conceição, M. J. A. Wood, I. Mäger and T. C. Roberts, RNA Biol., 2019, 16, 696–706.
31. G. Tzimagiorgis, E. Z. Michailidou, A. Kritis, A. K. Markopoulos and S. Kouidou, Cancer Epidemiol., 2011, 35, 580–589.
32. A. Baldassarre, C. Felli, G. Prantera and A. Masotti, Genes, 2017, 8, 234.
33. J. Shen, S. A. Stass and F. Jiang, Cancer Lett., 2013, 329, 125–136.
34. M. de Planell-Saguer and M. C. Rodicio, Anal. Chim. Acta, 2011, 699, 134–152.
35. G. Cheng, Adv. Drug Delivery Rev., 2015, 81, 75–93.
36. M. A. Javidi, A. H. Ahmadi, B. Bakhshinejad, N. Nouraee, S. Babashah and M. Sadeghizadeh, Med. Oncol., 2014, 31, 295.
37. J. A. Weber, D. H. Baxter, S. Zhang, D. Y. Huang, K. H. Huang, M. J. Lee, D. J. Galas and K. Wang, Clin. Chem., 2010, 56, 1733–1741.
38. C. Thery, F1000 Biol. Rep., 2011, 3, 15.
39. X. Chen, Y. Ba, L. Ma, X. Cai, Y. Yin, K. Wang, J. Guo, Y. Zhang, J. Chen, X. Guo, Q. Li, X. Li, W. Wang, Y. Zhang, J. Wang, X. Jiang, Y. Xiang, C. Xu, P. Zheng, J. Zhang, R. Li, H. Zhang, X. Shang, T. Gong, G. Ning, J. Wang, K. Zen, J. Zhang and C.-Y. Zhang, Cell Res., 2008, 18, 997–1006.
40. H. Loeser, M. von Brandenstein, V. Burst, C. Richter, R. Buettner, C. Kurschat, B. Hoppe and J. W. U. Fries, J. Mol. Med. Clin. Appl., 2018, 2, 1–10.
41. K. Wright, K. de Silva, A. C. Purdie and K. M. Plain, Sci. Rep., 2020, 10, 825.
42. Y.-K. Kim, J. Yeo, B. Kim, M. Ha and V. N. Kim, Mol. Cell, 2012, 46, 893–895.
43. T. X. Lu and M. E. Rothenberg, J. Allergy Clin. Immunol., 2018, 141, 1202–1207.
44. F. Marabita, P. de Candia, A. Torri, J. Tegner, S. Abrignani and R. L. Rossi, Briefings Bioinf., 2016, 17, 204–212.
45. R. A. M. Brown, M. R. Epis, J. L. Horsham, T. D. Kabir, K. L. Richardson and P. J. Leedman, BMC Biotechnol., 2018, 18, 16.
46. A. Brunet-Vega, C. Pericay, M. E. Quílez, M. J. Ramírez-Lázaro, X. Calvet and S. Lario, Anal. Biochem., 2015, 488, 28–35.
47. M. Monleau, S. Bonnel, T. Gostan, D. Blanchard, V. Courgnaud and C.-H. Lecellier, BMC Genomics, 2014, 15, 395.
48. J. D. Arroyo, J. R. Chevillet, E. M. Kroh, I. K. Ruf, C. C. Pritchard, D. F. Gibson, P. S. Mitchell, C. F. Bennett, E. L. Pogosova-Agadjanyan, D. L. Stirewalt, J. F. Tait and M. Tewari, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 5003–5008.
49. K. C. Vickers, B. T. Palmisano, B. M. Shoucri, R. D. Shamburek and A. T. Remaley, Nat. Cell Biol., 2011, 13, 423–433.
50. M. B. Kirschner, S. C. Kao, J. J. Edelman, N. J. Armstrong, M. P. Vallely, N. van Zandwijk and G. Reid, PLoS One, 2011, 6, e24145.
51. J. S. McDonald, D. Milosevic, H. V. Reddi, S. K. Grebe and A. Algeciras-Schimnich, Clin. Chem., 2011, 57, 833–840.
52. A. Garcia-Elias, L. Alloza, E. Puigdecanet, L. Nonell, M. Tajes, J. Curado, C. Enjuanes, O. Díaz, J. Bruguera, J. Martí-Almor, J. Comín-Colet and B. Benito, Sci. Rep., 2017, 7, 7725.
53. I. Moret, D. Sánchez-Izquierdo, M. Iborra, L. Tortosa, A. Navarro-Puche, P. Nos, J. Cervera and B. Beltrán, PLoS One, 2013, 8, e82753.
54. C. Becker, A. Hammerle-Fickinger, I. Riedmaier and M. W. Pfaffl, Methods, 2010, 50, 237–243.
55. M. Mraz, K. Malinova, J. Mayer and S. Pospisilova, Biochem. Biophys. Res. Commun., 2009, 390, 1–4.
56. P. Mestdagh, N. Hartmann, L. Baeriswyl, D. Andreasen, N. Bernard, C. Chen, D. Cheo, P. D’Andrade, M. DeMayo, L. Dennis, S. Derveaux, Y. Feng, S. Fulmer-Smentek, B. Gerstmayer, J. Gouffon, C. Grimley, E. Lader, K. Y. Lee, S. Luo, P. Mouritzen, A. Narayanan, S. Patel, S. Peiffer, S. Ruberg, G. Schroth, D. Schuster, J. M. Shaffer, E. J. Shelton, S. Silveria, U. Ulmanella, V. Veeramachaneni, F. Staedtler, T. Peters, T. Guettouche, L. Wong and J. Vandesompele, Nat. Methods, 2014, 11, 809–815.
57. H. Ravan, S. Kashanian, N. Sanadgol, A. Badoei-Dalfard and Z. Karami, Anal. Biochem., 2014, 444, 41–46.
58. W. Zhou, Y.-F. Tian, B.-C. Yin and B.-C. Ye, Anal. Chem., 2017, 89, 6120–6128.
59. F. Hu, J. Xu and Y. Chen, Anal. Chem., 2017, 89, 10071–10077.
60. D. Zubakov, A. W. M. Boersma, Y. Choi, P. F. van Kuijk, E. A. C. Wiemer and M. Kayser, Int. J. Leg. Med., 2010, 124, 217–226.
61. Y. Chen, J. A. L. Gelfond, L. M. McManus and P. K. Shireman, BMC Genomics, 2009, 10, 407.
62. K. A. Cissell, S. Shrestha and S. K. Deo, Anal. Chem., 2007, 79, 4754–4761.
63. T. Tian, J. Wang and X. Zhou, Org. Biomol. Chem., 2015, 13, 2226–2238.
64. I. Krepelkova, T. Mrackova, J. Izakova, B. Dvorakova, L. Chalupova, R. Mikulik, O. Slaby, M. Bartos and V. Ruzicka, Biotechniques, 2019, 66, 277–284.
65. C. Chen, D. A. Ridzon, A. J. Broomer, Z. Zhou, D. H. Lee, J. T. Nguyen, M. Barbisin, N. L. Xu, V. R. Mahuvakar, M. R. Andersen, K. Q. Lao, K. J. Livak and K. J. Guegler, Nucleic Acids Res., 2005, 33, e179.
66. K. Lao, N. L. Xu, V. Yeung, C. Chen, K. J. Livak and N. A. Straus, Biochem. Biophys. Res. Commun., 2006, 343, 85–89.
67. P. Mestdagh, T. Feys, N. Bernard, S. Guenther, C. Chen, F. Speleman and J. Vandesompele, Nucleic Acids Res., 2008, 36, e143.
68. M. Fauth, A. B. Hegewald, L. Schmitz, D. J. Krone and M. J. Saul, FASEB BioAdv., 2019, 1, 481–492.
69. Codex Committee on Methods of Analysis and Sampling, CAC/GL 74-2010. Guidelines on performance criteria and validation of methods for detection, identification and quantification of specific DNA sequences and specific proteins in foods, 2010.
70. European Network of GMO Laboratories, JRC Technical Report: Definition of Minimum Performance Requirements for Analytical Methods of GMO Testing, 2015.
71. S. A. Bustin, V. Benes, J. A. Garson, J. Hellemans, J. Huggett, M. Kubista, R. Mueller, T. Nolan, M. W. Pfaffl, G. L. Shipley, J. Vandesompele and C. T. Wittwer, Clin. Chem., 2009, 55, 611–622.
72. S. Broeders, I. Huber, L. Grohmann, G. Berben, I. Taverniers, M. Mazzara, N. Roosens and D. Morisset, Trends Food Sci. Technol., 2014, 37, 115–126.
73. J. R. Dijkstra, L. J. M. Mekenkamp, S. Teerenstra, I. De Krijger and I. D. Nagtegaal, J. Cell. Mol. Med., 2012, 16, 683–690.
74. H. J. Peltier and G. J. Latham, RNA, 2008, 14, 844–852.
75. K. H. Chang, P. Mestdagh, J. Vandesompele, M. J. Kerin and N. Miller, BMC Cancer, 2010, 10, 173.
76. J. J. Gooding and K. Gaus, Angew. Chem., Int. Ed., 2016, 55, 11354–11366.
77. M. A. Morales and J. M. Halpern, Bioconjugate Chem., 2018, 29, 3231–3239.
78. X. Zhang, Q. Guo and D. Cui, Sensors, 2009, 9, 1033–1053.
79. B. Lu, L. Liu, J. Wang, Y. Chen, Z. Li, S. C. B. Gopinath, T. Lakshmipriya and Z. Huo, Nanoscale Res. Lett., 2020, 15, 105.
80. S. Mathur, A. Erdem, C. Cavelius, S. Barth and J. Altmayer, Sens. Actuators, B, 2009, 136, 432–437.
81. W. Putzbach and N. J. Ronkainen, Sensors, 2013, 13, 4811–4840.
82. I. Gessner and S. Mathur, in Nanotechnologies in Preventive and Regenerative Medicine, ed. V. Uskovic, Elsevier Ltd, Oxford, 1st edn, 2018, pp. 260–297.
83. N. M. Green, Biochem. J., 1963, 89, 599–609.
84. F. Degliangeli, P. P. Pompa and R. Fiammengo, Chem. – Eur. J., 2014, 20, 9476–9492.
85. K. M. Koo, N. Soda and M. J. A. Shiddiky, Curr. Opin. Electrochem., 2020, 100645.
86. S. Berensmeier, Appl. Microbiol. Biotechnol., 2006, 73, 495–504.
87. S. S. Leong, S. P. Yeap and J. K. Lim, Interface Focus, 2016, 6, 20160048.
88. J. H. Choi, J. H. Lee, J. Son and J. W. Choi, Sensors, 2020, 20, 1003.
89. J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and R. P. Van Duyne, Nanosci. Technol. A Collect. Rev. from Nat. Journals, 2009, 7, 308–319.
90. L. Shi, Y. Sun, L. Mi and T. Li, ACS Sens., 2019, 4, 3219–3226.
91. J. Ki, H. Y. Lee, H. Y. Son, Y. M. Huh and S. Haam, ACS Appl. Mater. Interfaces, 2019, 11, 18923–18929.
92. S. Shin Low, Y. Pan, D. Ji, Y. Li, Y. Lu, Y. He, Q. Chen and Q. Liu, Sens. Actuators, B, 2020, 308, 127718.
93. N. Fakhri, S. Abarghoei, M. Dadmehr, M. Hosseini, H. Sabahi and M. R. Ganjali, Spectrochim. Acta, Part A, 2020, 227, 117529.
94. Y. Sun and T. Li, Anal. Chem., 2018, 90, 11614–11621.
95. Z. Wang, Z. Xue, X. Hao, C. Miao, J. Zhang, Y. Zheng, Z. Zheng, X. Lin and S. Weng, Anal. Chim. Acta, 2020, 1103, 212–219.
96. Y. S. Borghei, M. Hosseini and M. R. Ganjali, J. Photochem. Photobiol., A, 2020, 391, 112351.
97. Y. Y. Bai, Z. Wu, C. M. Xu, L. Zhang, J. Feng, D. W. Pang and Z. L. Zhang, Anal. Chem., 2020, 92, 853–959.
98. X. Cao, K. Zhang, W. Yan, Z. Xia, S. He, X. Xu, Y. Ye, Z. Wei and S. Liu, Microchim. Acta, 2020, 187, 212.
99. W. Y. Chen, C. H. Wang, K. H. Wang, Y. L. Chen, L. K. Chau and S. C. Wang, Biomicrofluidics, 2020, 14, 014102.
100. Y. Pang, C. Wang, J. Wang, Z. Sun, R. Xiao and S. Wang, Biosens. Bioelectron., 2016, 79, 574–580.
101. L. Yu, P. He, Y. Xu, X. Kou, Z. Yu, X. Xie and P. Miao, Sens. Actuators, B, 2020, 313, 128015.
102. P. Chen, X. Jiang, K. Huang, P. Hu, X. Li, L. Wei, W. Liu, L. Wei, C. Tao, B. Ying, X. Wei and J. Geng, ACS Appl. Mater. Interfaces, 2019, 11, 36476–36484.
103. K. Yang, L. Feng, X. Shi and Z. Liu, Chem. Soc. Rev., 2013, 42, 530–547.
104. C. Zhang, P. Miao, M. Sun, M. Yan and H. Liu, Small, 2019, 15, 1–23.
105. M. S. Hizir, M. Balcioglu, M. Rana, N. M. Robertson and M. V. Yigit, ACS Appl. Mater. Interfaces, 2014, 6, 14772–14778.
106. Q. Xi, D. M. Zhou, Y. Y. Kan, J. Ge, Z. K. Wu, R. Q. Yu and J. H. Jiang, Anal. Chem., 2014, 86, 1361–1365.
107. B. Cai, S. Guo and Y. Li, Anal. Methods, 2018, 10, 230–236.
108. M. N. Islam, L. Gorgannezhad, M. K. Masud, S. Tanaka, M. S. A. Hossain, Y. Yamauchi, N.-T. Nguyen and M. J. A. Shiddiky, ChemElectroChem, 2018, 5, 2488–2495.
109. S. Bi, M. Chen, X. Jia and Y. Dong, Nanoscale, 2015, 7, 3745–3753.
110. M. Daneshpour, B. Karimi and K. Omidfar, Biosens. Bioelectron., 2018, 109, 197–205.
111. M. Oishi, ACS Omega, 2018, 3, 3321–3329.
112. E. Hosseinzadeh, H. Ravan, A. Mohammadi and R. Mohammad-rezaei, Biosens. Bioelectron., 2018, 117, 567–574.
113. X. Li, J. Zhao, R. Xu, L. Pan and Y. M. Liu, Analyst, 2020, 145, 1783–1788.
114. A. A. Svoronos, D. M. Engelman and F. J. Slack, Cancer Res., 2016, 76, 3666–3670.
115. Y. H. Feng and C. J. Tsao, Biomed. Rep., 2016, 5, 395–402.
116. S. Li, K. He, R. Liao, C. Chen, X. Chen and C. Cai, Talanta, 2017, 174, 679–683.
117. S. Tang, Y. Li, A. Zhu, Y. Yao, J. Sun, F. Zheng, Z. Lin and W. Shen, Chem. Commun., 2019, 55, 8386–8389.
118. B. Tian, J. Ma, Z. Qiu, T. Zardán Gómez de la Torre, M. Donolato, M. F. Hansen, P. Svedlindh and M. Strömberg, ACS Nano, 2017, 11, 1798–1806.
119. H. Chan, S. Ho, D. He and H. Li, Talanta, 2020, 206, 120217.
120. Z. Shen, L. He, W. Wang, L. Tan and N. Gan, Biosens. Bioelectron., 2020, 148, 111831.
121. T. Qi, C. Song, J. He, W. Shen, D. Kong, H. Shi, L. Tan, R. Pan, S. Tang and H. K. Lee, Anal. Chem., 2020, 92, 5033–5040.
122. J. J. Wang, C. Zheng, Y. Z. Jiang, Z. Zheng, M. Lin, Y. Lin, Z. L. Zhang, H. Wang and D. W. Pang, Anal. Chem., 2020, 92, 830–837.
123. R. Tavallaie, J. McCarroll, M. Le Grand, N. Ariotti, W. Schuhmann, E. Bakker, R. D. Tilley, D. B. Hibbert, M. Kavallaris and J. J. Gooding, Nat. Nanotechnol., 2018, 13, 1066–1071.
124. N. Zhang, X. M. Shi, H. Q. Guo, X. Z. Zhao, W. W. Zhao, J. J. Xu and H. Y. Chen, Anal. Chem., 2018, 90, 11892–11898.
125. R. Chand, S. Ramalingam and S. Neethirajan, Nanoscale, 2018, 10, 8217–8225.
126. C. Mall, D. M. Rocke, B. Durbin-Johnson and R. H. Weiss, Biomarkers Med., 2013, 7, 623–631.
127. S. Pospisilova, E. Pazourkova, A. Horinek, A. Brisuda, I. Svobodova, V. Soukup, J. Hrbacek, O. Capoun, T. Hanus, J. Mares, M. Korabecena and M. Babjuk, Neoplasma, 2016, 63, 799–808.
128. J. W. U. Fries, Biol. Cell., 2019, 111, 169–186.
129. S. Xu, S. Hossaini Nasr, D. Chen, X. Zhang, L. Sun, X. Huang and C. Qian, ACS Biomater. Sci. Eng., 2018, 4, 654–662.
130. X. Huang, Z. Xu, J. H. Liu, B. Y. Yu and J. Tian, RSC Adv., 2020, 10, 11257–11262.
131. W. Lu, Y. Chen, Z. Liu, W. Tang, Q. Feng, J. Sun and X. Jiang, ACS Nano, 2016, 10, 6685–6692.
132. M. Z. Mousavi, H. Y. Chen, K. L. Lee, H. Lin, H. H. Chen, Y. F. Lin, C. S. Wong, H. F. Li, P. K. Wei and J. Y. Cheng, Analyst, 2015, 140, 4097–4104.
133. N. Soda, B. H. A. Rehm, P. Sonar, N. T. Nguyen and M. J. A. Shiddiky, J. Mater. Chem. B, 2019, 7, 6670–6704.
134. J. Lim, M. Choi, H. Lee, Y. H. Kim, J. Y. Han, E. S. Lee and Y. Cho, J. Nanobiotechnol., 2019, 17, 1–12.

---