Resonance Energy Transfer-Based Biosensors for Point-of-Need Diagnosis-Progress and Perspectives

Abstract

The demand for point-of-need (PON) diagnostics for clinical and other applications is continuing to grow. Much of this demand is currently serviced by biosensors, which combine a bioanalytical sensing element with a transducing device that reports results to the user. Ideally, such devices are easy to use and do not require special skills of the end user. Application-dependent, PON devices may need to be capable of measuring low levels of analytes very rapidly, and it is often helpful if they are also portable. To date, only two transduction modalities, colorimetric lateral flow immunoassays (LFIs) and electrochemical assays, fully meet these requirements and have been widely adopted at the point-of-need. These modalities are either non-quantitative (LFIs) or highly analyte-specific (electrochemical glucose meters), therefore requiring considerable modification if they are to be co-opted for measuring other biomarkers. Förster Resonance Energy Transfer (RET)-based biosensors incorporate a quantitative and highly versatile transduction modality that has been extensively used in biomedical research laboratories. RET-biosensors have not yet been applied at the point-of-need despite its advantages over other established techniques. In this review, we explore and discuss recent developments in the translation of RET-biosensors for PON diagnoses, including their potential benefits and drawbacks.

Keywords: BRET; CRET; FRET; PADs; microfluidics; on-site; on-the-spot; point-of-care; time-resolved FRET.

Figure 1

(a) Overview of different variations on the Resonance Energy Transfer principle, together with a range of different types of biological recognition element. Red domains indicate recognition elements, and orange domains indicate targeted analytes. (b) Examples of different ways that energy donors and acceptors can be combined with biological recognition elements. All illustrations are simplified and not shown to scale. Images were taken from the following sources: The “Fluorescent proteins” image illustrates the Green Fluorescent Protein (doi:10.2210/rcsb_pdb/mom_2003_6). The “Organic dyes” structure shows cyanine. The quantum dot image was taken from Reference [49]. The “Luciferase–Luciferin” image is composed of the Firefly luciferase (doi:10.2210/rcsb_pdb/mom_2006_6) and D-Luciferin as its substrate. The structure of horseradish peroxidase was taken from the Protein Data Bank (1W4E, doi:10.2210/pdb1W4W/pdb). The dark quencher is the black hole quencher BHQ1 from atdbio (https://www.atdbio.com/content/35/FRET-fluorescence-quenchers).

Figure 2

Examples of point-of-need (PON)-suitable applications using compact RET detection devices in combination with microfluidics. In one example, Bioluminescence Resonance Energy Transfer (BRET)-based biosensors are run on a microfluidics chip integrated into a compact device containing micro photon multiplier tubes (µPMTs). In other examples, FRET biosensor signals are recorded by using a fluorescence microscope or laser excitation followed by detection using PMTs. (a) Lymphocytes secreting Matrix metalloproteinase-9 (MMP9) are trapped by antibodies in a micro well located on a microfluidic chip. The peptides are labeled with Fluorescein isothiocyanate (FITC) and 4-(dimethylaminoazo)benzene-4-carboxylic acid (DABCYL), and they contain MMP9-specific cleavage sites. These are immobilized close to the micro wells, to detect any MMP9 activity released by the cells [52]. (b) Aptamers, labeled with quantum dots (QDs), specific for cancer-related cells or protein markers are attached on a graphene monoxide layer. Binding of the target cells or proteins to their specific aptamer results in a release of the aptamer from graphene oxide, activating the fluorescence signal of the quantum dot/organic dye [53,54]. (c) CYBERTONGUE^® protease biosensors consist of the Renilla luciferase RLuc8 connected through a peptide linker, containing specific recognition sites for the target protease, to the Green Fluorescent Protein variant GFP². Proteolytic activity exerted on the connecting peptide results in the dissociation of GFP² from RLuc8, leading to a profound change in BRET ratio [48,55,56]. (d) The CYBERTONGUE^® lactose biosensor consists of a lactose-binding protein tagged with RLuc8 and GFP² that undergoes a conformational change upon binding to lactose [57]. Binding of lactose results in the distancing of the two BRET components, thereby changing the BRET ratio.

Figure 3

Overview of the Cybertongue^® BRET analysis device: (a) functional schematic of the measurement device, (b) schematic design of a microfluidic chip used for protease assays and (c) image of compact microfluidics device with closed lid. Figure was taken from Weihs et al. [48], with permission.

Figure 4

Examples of PON-suitable applications using a digital camera or smartphone in combination with micro plates or paper-based devices. BRET-based biosensors are spotted on paper-based analytical devices (PADs), and signals are recorded with a digital camera or smart phone. FRET biosensors require an additional source of excitation, such as a light-emitting diode (LED) or UV-lamp. (a) LUMABS biosensors (LUMinescent AntiBody Sensor) are comprised of the luciferase NanoLuc and the fluorescent protein mNeonGreen connected through a linker containing linear epitopes for antibodies of interest. In the absence of the antibody of interest, NanoLuc and mNeonGreen dimerize through the connector domains [73,74,78]. (b) LUMABs were modified by replacing linear epitopes with unnatural amino acids acting as a chemical handle to introduce analogues of analytes of interest. An antibody binding to these analogues is introduced, separating NanoLuc and mNeonGreen. In the presence of the analyte, antibodies preferentially bind to the analyte instead of its analogues incorporated in the LUMAB biosensor [77]. (c) LUCIDs (luciferase-based indicators of drugs) are protein fusions comprising NanoLuc, a receptor protein for the drug of interest and the self-labeling enzyme SNAP. A SNAP-functionalized organic dye Cy3 is attached to an analyte analogue, which is covalently incorporated by the SNAP protein. In the presence of the analyte, the receptor preferentially binds the analyte over its SNAP–Cy3-bound analogue [79,80,81]. (d) If target miRNAs are present in a sample, miRNA templates are amplified through a rolling circle amplification (not illustrated). Complementary oligonucleotides form double-stranded DNAs that are recognized by fusions of NanoLuc and mNeonGreen with zinc finger proteins that specifically bind to different but nearby sequences [82]. (e) Quantum dot (QD)–organic fluorescent dye conjugates joined by a peptide-containing protease-specific recognition site are immobilized on paper. In the absence of proteolytic activity, FRET occurs between the QD and the organic dye, resulting in a yellow/orange emission signal. If the peptide is cleaved due to the proteolytic activity of the protease of interest, the dye diffuses away from the QD, leading to a green emission from the QDs [83]. (f) Paper-immobilized quantum dot–oligonucleotides and free Cy3–oligonucleotides contain different DNA segments complementary to the target gene fragment. In a sandwich format, the target gene serves as a hybridization bridge for the QD–oligonucleotide and Cy3–oligonucleotide, which in turn enables FRET between QD and Cy3 [84,85,86,87]; (g) A Cy3-labeled kanamycin-specific aptamer partially hybridizes to an anchor/connector oligonucleotide immobilized on glass. The connector oligonucleotide is conjugated to Cy5. Binding of kanamycin spatially separates Cy3 from Cy5 components, leading to a lower FRET efficiency [88]. (h) An upconversion nanoparticle (UCNP) consisting of ytterbium (Yb³⁺) and thulium (Tm³⁺) is conjugated to the organic dye rhodol. FRET occurs between the UCNP and rhodol, while organophosphonates perform a nucleophilic attack on rhodol, inactivating it as a FRET acceptor [89].

Figure 5

Examples of Lanthanide-FRET (LRET) applications that are potentially suitable for on-site testing. (a) Commonly applied TR-FRET technologies rely on a sandwich-based homogeneous assay, where two antibodies targeting different epitopes of the analyte are labeled with a lanthanide energy donor, while the other is labeled, with an organic dye or fluorescent protein, as the energy acceptor. Binding of both antibodies enables FRET between the lanthanide and acceptor which is measured by their altered fluorescence lifetimes. (b) Protein L, an antibody light-chain-binding protein [102], is labeled with Europium. Antigens to an antibody of interest are labeled with the organic dye AlexaFluor647 (LFRET) [103]. If a sample contains antibodies against the antigen–dye fusion, FRET occurs between Protein-L-Europium bound to the light chain of the antibody and the Antigen–AlexaFluor647 fusion. Image of the ProciseDx device was used with permission from ProciseDx.