The use of nanocrystal quantum dot as fluorophore reporters in molecular beacon-based assays

2.1 Factors influencing MB operation

There are several factors that influences the operational performance of MB-based probes. The distance between the quencher-fluorophore pair is the primary factor that determines the quenching efficiency of the probe [21]. Based on the concept of FRET, the energy transfer efficiency is proportional to the 6th power of the distance between the donor–acceptor pair. Studies have shown that efficient energy transfer can yield 10–100-fold increase in fluorescence intensity upon hybridization [2].

Another factor which influences MB operation is temperature. At low temperature, a stable hairpin structure of the MB can be maintained but distortion of the structure can occur at higher temperature [29]. The chain length of the stem influences the melting temperature, ionic strength of the buffer and on the guanine-cytosine content. Bonnet et al. reported the effect of temperature on a MB with perfect complementary sequence [21]. At low temperature, a complementary duplex hybrid between the loop sequence of the MB and the target sequence was formed, thus generating strong fluorescence recovery signal. As the temperature was gradually increased, the duplex hybrid formed between the MB and the target sequence was broken and the MB returned to its original hairpin state resulting in a weakened fluorescence signal. With further increase in the temperature, the hairpin structure of the MB was completely destroyed, resulting in a linear structure and restoration of the fluorescence.

The pH of the MB probe solution is also a factor that influences its performance. At high pH, the stem of the MB is broken and this degrades the MB leading to a false negative signal [2].

2.2 MB synthesis

The synthesis MB mimics the synthesis of dual-labeled oligomers with two dyes [30]. The loop sequence length (15–40 bp nucleotide) is selected so as to ensure a stable target-probe hybrid at the chosen temperature. The choice of the stem length (5–7 bp nucleotide) should be appropriated such as it forms a stable hairpin stem-loop structure for effective fluorescence quenching and should also be weak enough to dissociate when a duplex hybrid is formed between the probe and the target sequence.

The most commonly used quencher is DABCYL, known as a universal quencher due to its ability to quench a wide range of fluorescent-emitting fluorophores [2]. Hence, it is usually used as a starting material for MB synthesis via a “DABCYL-controlled pore glass (CPG)” [30]. A variety of fluorescent dye molecules of choice can be directly linked to the 5′ end of the probe to report the fluorescence signal based on the structural change in probe structure.

There are four main steps involved in the synthesis of MBs [30]; Firstly, DABCYL is derivatized with the CPG solid support and used to initiate the synthesis at the 3′ end. Afterward, sequential addition of the rest of the nucleotide is performed using standard cyanoethylphosphoramidite technique. Secondly, at the 5′end, a primary amine group is linked. Thirdly, hydrolysis of the oligonucleotide is performed followed by removal from the CPG and finally purified by reverse-phase liquid chromatography. Lastly, labelling of the purified oligonucleotide with a fluorophore can be performed and the excess fluorophore dye is then removed by gel chromatography on a Sephadex G-25 column or a Qdot fluorophore can be directly conjugated to the MB [27].

2.3 Advantages of MB over other nucleic acid probes

Several DNA fluorescence probes have employed the hybridization technology as a means to target a complementary sequence. Nucleic acid blotting assay is a classic example of techniques employing hybridization technology. This technique has contributed to the understanding of gene technology. The technique operates by using a solid support to incorporate DNA fragments and an oligonucleotide probe embedded with the target sequence is used to aid hybridization [30, 31]. However, for real-time monitoring of nucleic acid synthesis and nucleic acid labeling in living cells, nucleic acid blotting assays, or probes using target-probe isolation or other assays using intercalating reagents cannot be used for such applications.

Fluorescence transduction signal mechanism, which is influenced by changes within the chemical environment of a fluorophore molecule, has enabled MB to function as ultrasensitive probes for real-time monitoring whilst generating high background signal-to-noise ratio. The fluorescence “switch ON” state of a MB can increase up to 200-fold when hybridized with the target nucleic acid [2]. Hence, with respect to signal generation, MB holds the advantage over other oligonucleotide probes. This is corroborated by the imaging of single molecule DNA in living system [32]. MB possesses the functionality of being effective in situation where it is practically difficult or undesirable to separate the target-probe hybrid from an excess of unhybridized solution. This is essentially needed in the detection of mRNAs in living cells or in real-time PCR in sealed tubes [30]. We cannot overemphasize the degree of usefulness of MB in detection without the separation of the target-probe. This process enables the direct monitoring of nucleic acids during synthesis in living samples or in sealed tubes, without the need for further manipulation.

The molecular specificity of MB is another unique advantage. MBs are extremely target specific and are practically effective in single nucleotide polymorphism (SNP) detection. Conventional technique for SNP detection is time consuming and labor-intensive but MBs offer a simple and direct means for genetic disease diagnosis and in gene therapy studies [5]. Studies have shown that hybridization from perfect complimentary DNA sequence and nonhybridization from mismatch DNA sequence occur at a wider temperature for MB probes than for traditional oligonucleotide linear probes [21].

3.1 Comparison of the optical properties of Qdots and fluorescent organic dyes

Organic fluorescent dyes can be classified as traditional fluorophores used in MB biosensor design. Subsequently, fluorescent Qdots came into the fray and have attracted interest due to their unique optical properties and biocompatibility. Organic fluorescent dyes are generally bonded to the MB during synthesis whilst Qdots nanocrystals are directly conjugated to the MB in solution. In this section, we compare the optical properties of semiconductor Qdot nanocrystals with those of organic fluorescent dyes.

It is important to emphasize that the optical properties of Qdots arise from interactions between hole and electron within their local chemical environments. The chemistry associated with the fluorescence of Qdots involves the excitation of electron from the valence band to the conduction band, hence creating a hole in the valence band. With respect to the emission line width, those of Qdots are narrower than organic fluorescent dyes. The extent of narrowing or broadening of the emission line width provides information on the quality of size distribution of the materials as well as surface defect states. Hence, narrower emission line width is more favorable and Qdots holds the advantage. Organic fluorescent dyes are highly prone to photobleaching due to degradation of their fluorescence after prolonged exposure to light, whereas Qdots are ~100–200 times more stable and ~20 more brighter [47]. It is estimated that the molar extinction coefficient of Qdots (CdSe), which depends on the excitation wavelength and particle size, range between 10⁵ and 10⁶ M⁻¹ cm⁻¹ and is 10–100 times larger than those of organic fluorescent dyes [48]. The quantum size effect of Qdots is a unique advantage over organic fluorescent dyes due to the ability to obtain several Qdots size of the same structure and with varying optical properties, thus making them suitable for multiplex detection and the ability to select appropriate Qdot size for the targeted application.

3.2 The chemistry and optimization strategies of Qdots-MB probes

A comparative study was carried out by Kim et al. in [49] in which mercaptoacetic acid (MAA)-capped CdSe/ZnS Qdots was conjugated to the 5′ end of the MB with DABCYL as the fluorescence quencher and alternatively, organic fluorescent dye, 6-FAM was attached to the 5′end, also using DABCYL as the quencher molecule. The interaction between the Qdot or 6-Fam with DABCYL was attributed to FRET, hence the chemical interaction was studied on this basis. A more extended overlap of the fluorescence of the Qdot (90%) with the absorption spectrum of DABCYL was observed over that of 6-Fam (30%). Using molecular modeling simulations, the potential distance between the Qdot/DABCYL and 6-Fam/DABCYL was determined. Figure 3 shows the snap shot of the potential distance (nearest and farthest) between Qdot/DABCYL and 6-Fam/DABCYL (nearest and farthest). It was predicted that the potential distance, nearest and farthest between the Qdot and DABCYL was 3.3 and 5 nm, respectively. The distance was simulated from the N=N bond (center) of DABCYL to the center of the Qdot. The Förster resonance distance was calculated to be 3.4 nm while the FRET efficiency for the farthest and nearest Qdot/DABCYL distance was calculated to be ~9 and ~54%, respectively. The potential nearest and farthest distance between DABCYL and 6-Fam was calculated to be 2.4 and 3.4 nm which was relatively smaller than that for the Qdot/DABCYL distance.

The distance measurement of 6-Fam/DABCYL was measured from the center of 6-Fam (C9 in Xanthene) to the N=N bond of the quencher, DABCYL. The Förster distance was calculated to be 4.6 nm while the FRET efficiency was calculated to be 98%. It was observed that the larger center-to-center spacing between DABCYL and the Qdot was relatively larger than that of 6-Fam/DABCYL. This difference was attributed to the larger size effect of the Qdots which also influenced its FRET efficiency of the Qdots and the Förster radius between the Qdot and DABCYL. Since this study was limited to a single Qdot size, the quantum size effect of Qdots which could yield several sizes across the visible region to the near infra-red will have a profound effect on the interaction with quencher molecules in MB assays.

In another related development, the optimized linkage and quenching effects of different quencher molecule to the MB were studied by Cady et al. when using Qdot as the fluorophore reporter [26]. In their study, different quenchers, DABCYL, Iowa Black and 1.4 nm gold nanoparticles, bonded to the MB was used as a means to study the quenching interaction with the Qdot fluorophore. Carboxyl-modified Qdots were conjugated to the amino-linked 5′end of Iowa Black and DABCYL bonded MBs while streptavidin-functionalized Qdots were bonded to biotinylated MBs with gold nanoparticle and Iowa Black as quencher molecules. It was observed that Iowa Black exhibited a similar quenching efficiency to gold nanoparticle and both were better than DABCYL. With respect to the linkage strategies of the Qdots to the MBs, it was observed that carboxyl-bonded MBs generated better output efficiency than streptavidin-bonded MBs. The difference was attributed to the binding efficiency between the numbers of accessible streptavidin or carboxyl groups on the Qdots. It was suggested that the secondary hairpin structure formation could be impeded by the packing of MB DNA on the Qdots surface and thereafter results in low quenching efficiency of the Qdot (in the absence of the target nucleic acid) and small fluorescence enhancement changes upon hybridization with the target. The relative distance between the quencher molecules and the fluorophore reporter was also suggested to influence the overall performance of the MB probe. In terms of FRET efficiency, carboxyl-functionalized Qdot exhibited an 89% energy transfer rate in comparison to 14% exhibited by streptavidin-functionalized Qdots.

4.1 DNA detection

The detection of DNA is by far the most common application of Qdot-MB probes. Cady et al. developed a Qdot-MB biosensor for DNA [26] and studied the fluorescence enhancement efficiency based on different linkage strategies and different fluorescence quenchers bonded to the MB. Carboxyl-modified Qdots bonded to Iowa Black or DABCYL-bonded MB and streptavidin-modified Qdots bonded to Iowa Black and gold nanoparticle-bonded MB were used in their study. They observe that the fluorescence increase after hybridization with the target DNA followed the order Carboxyl-Qdot/Iowa Black MB (3.3-fold increase) > streptavidin-Qdot/Iowa Black MB (2.1-fold increase) > streptavidin-Qdot/gold nanoparticle MB (1.9-fold increase) > Carboxyl-Qdot/DABCYL MB (1.0-fold increase).

A novel strategy for MB binding readout for DNA detection involving the separation of the Qdot signal reporter and the molecular recognition element was developed by Wang et al. [50]. A label-free MB-based DNA probe was developed by using an unmodified thymine-rich DNA oligonucleotide probe that is specific to the target DNA and also acting as the captured probe. Hg²⁺ which served as the competitor, binds partially to the capture DNA oligonucleotide signal transducer while the Qdot reporter (MAA-capped CdTe/ZnS) was used as the signal reporter. In the absence of the target DNA, a hairpin structure is formed upon binding of Hg²⁺ to the capture DNA probe. Upon interaction of the target DNA, displacement of the bound Hg²⁺ from the hairpin structure results in binding interaction with the Qdots and the signal read-out is obtained from the Qdot fluorescence. A detection limit of 25 nM for DNA detection was reported. Compared with conventional MB, the label-free MB was proposed to exhibit three notable features; (i) by separating the Qdot signal reporter and the molecular recognition element, the double-labeled formation which could lead to lose of specificity and affinity of the oligonucleotide can be avoided; (ii) release of several Hg²⁺ to interact with the fluorophore reporter can lead to amplified detection signal and (iii) flexibility in the specificity and sensitivity adjustment can be achieved by the Hg²⁺-dependent stem formation. Point (iii) is quite unique because Hg²⁺ was used to modulate the sensitivity and specificity of the MB while for conventional MB probe, optimization of the specificity and sensitivity can be achieved by varying the length of the stem and loop sequence and this will require synthesizing several MBs.

A MB microarray assay for label free detection of DNA based on streptavidin-Qdot labeling on the MB instead of the target was developed by Guo et al. [51]. As shown in Fig. 4, their biosensor operates by functionalizing the MB with biotin and a thiol rather than a quencher and a fluorophore at the 3′ end and 5′end of the probe. The microarray was developed by immobilizing the MBs via the epoxy group on the surface of the slide and the 5′end of the terminal thiol group on the MB. In the presence of the target DNA, the hybridization effect triggers the opening of the MB, hence the streptavidin-Qdots label the MB via the biotin-steptavidin interaction. As shown in the Fig. 4, MB1 and MB2 were immobilized on one side of the microarray system. The microarray contained four subarrays in which four genotypes were acquired by mixing two mutant types and two wild types. The single nucleotide polymorphism (SNP) detection was carried out for the four genotypes on the four subarrays. A detection limit as low as 0.1 pM was reported.

A capillary electrophoresis-assisted GSH-CdTe Qdot-MB biosensor was developed for dual single-base DNA mutation detection by Li et al. [52]. To detect the dual single-base mutation, the 15-base pair oligonucleotide loop sequence of the Qdot-MB1 was complementary to the 3′end of the DNA target in which mutation existed at the G position while the 15-base pair loop sequence of the Qdot-MB2 was complementary to the 5′end of the DNA target in which the mutation existed at the T position. The hybridization process was monitored on a capillary electrophoretic device incorporated with fluorescence. A detection limit of 14 nM was reported.

Dong et al. [53] reported the detection of DNA using a 3-mercaptopropionic acid (MPA)-capped Qdot-MB biosensor probe in which graphene oxide (GO) was used as a fluorescence quencher. Firstly, the Qdot was bonded to the MB and the formed Qdot-MB probe was immobilized on GO. FRET from the Qdot to GO was proposed to be the mode of fluorescence quenching. Their work was the first demonstration of the use of a Qdot-MB probe in which traditional dark fluorescence quenchers were not used, rather a carbon-based nanomaterial was used to modulate the MB operation. The Qdot-based MB was used to detect perfect match, single-base mismatch and three-base mismatch DNA sequences. Their results showed that the detection of the perfect match sequence was 2.5 times higher than the single base mismatch and the fluorescence enhancement response of the three-base mismatch was only 15% that of the perfect complementary DNA target. A limit of detection of 12 nM was reported.

A silica-coated CdSe/ZnS Qdot-MB biosensor was developed by Wu et al. for the detection of DNA [54]. Both perfect complementary and single-base mismatch were detected. Their results showed that the detection of the perfect complementary DNA sequence was ten times more sensitive than the single-base mismatch for low sample DNA concentrations and 2–3 times more sensitive for high DNA concentrations.

β-lactamase gene present in antibiotic-resistance bacteria of plasmids pUC18 in E. coli strain DH5α was detected by Wu et al. using a MAA-capped QD-based MB [55]. Two types of MB hybridization strategies using BHQ2 as the quencher were developed, in which the first was a hairpin MB and the second was a double-stranded MB. Direct fluorescence in situ hybridization was used to monitor the hybridization between the target complementary DNA and the probe. The rate of movement of the band and its brightness were used to judge the fluorescence enhancement. It was reported that the hybridization between the target DNA and the Qdot-MB probe was reflected by the increase of the bands brightness. The increase of the fluorescence intensity upon hybridization was reported to be 3.8-fold.

4.2 MicroRNA detection

An ultrasensitive biosensor for microRNA involving the integration of a MB mediated rolling circle amplification (RCA) and using CdSeTe/CdS Qdot as a fluorophore tag was developed by Wang et al. [56] (Fig. 5). Specifically, miR-16 was selected as the target miRNA. They immobilized a locked nucleic acid (LNA) MB on the surface of a gold electrode and the target miRNA was interacted with the setup to open up the MB. This triggers the LNA-MB loop structure to hybridize with the target miRNA and thus release the RCA primer. Upon binding of the RCA template, Phi29 DNA polymerase and dNTPs were used to initiate the RCA reaction. They obtained a RCA product with thousands of repeated sequences of a long single-stranded DNA which was exploited for linear periodic hybridization when utilizing the Qdot detection probe. They proposed that the Qdot-based RCA product disintegrated into a random DNA coil in aqueous solution as evident from the fluorescence imaging. The low detection limit of 0.32 aM was reported.

Endogenous and exogenous miRNA fluorescence imaging using a Qdot-MB biosensor probe was developed by Lee et al. [57]. The cellular uptake was improved by binding a 9-mer arginine-rich peptide to the Qdots. miRNA124a which is known to be expressed in neurons was chosen as the target. In the absence of the target miRNA124a, the constructed R9-Qdot-miR124a-MB fluorescence was switched OFF while in the presence of the target miRNA124a, the fluorescence was switched ON. Their study is the only reported Qdot-MB biosensor for miRNA imaging.

4.3 Viral RNA detection

RNA-type viruses are an attractive target for Qdot-MB based probes due to the ability to design the loop to target a specific region within the virus RNA genome. It is important to emphasize that detection of viral RNA using MB technology requires the use of highly sensitive Qdot fluorophores. Our research group has recently exploited this technology to detect viral RNA of influenza virus and norovirus. We have reported on the development of a Qdot-MB biosensor probe for influenza virus RNA [46]. In the study, an ultrasensitive thioglycolic acid (TGA)-capped CdZnSeS/ZnSe_1.0S_1.3 alloyed Qdot with a fluorescence quantum yield of 98% and exhibiting unique brightness under UV light was synthesized. The synthesis of the Qdot was achieved via engineered band gap alloying. Viral RNA of influenza virus H1N1 in a low concentration range of 2–14 copies/mL was detected both in buffer solution and in human serum. A detection limit of 5.2 copies/mL was reported.

In another related development, an ultrasensitive biosensor for influenza virus H1N1 RNA was developed using a near-infrared quinternary MPA-CdZnSeTeS alloyed Qdot-MB [27]. This biosensor not only detected the viral RNA but it also discriminated between different strains of the virus, (A/Brisbane/59/2007) and (A/California/7/2009). The loop of the MB was complementary to 21 base pairs of (A/Brisbane/59/2007) strain and 22 base pairs of (A/California/7/2009). The Qdot-MB detected the viral RNA both in human serum and in buffer solution. A detection limit as low as ~1 copy/mL was reported.

Norovirus viral RNA detection has been reported using an ultrasensitive silica-functionalized alloyed CdZnSeS Qdot-MB biosensor probe [58]. The synthesis of highly fluorescent SiO₂-CdZnSeS Qdots as a fluorophore reporter bonded to the MB was achieved via a silanization process in which the surface of TGA-CdZnSeS Qdot was modified (Fig. 6). Norovirus RNA in the concentration range of 2–16 copies/mL was detected in buffer solution and 2–10 copies/mL in human serum (Fig. 7). A detection limit as low as 9.3 copies/mL was reported.

Finally, visualization of viral replication cycle using a self-assembled hexahistidine-anchored Tat peptide Qdot-MB-AuNP biosensor was developed by Yeh et al. [59]. Conventional CdSe/ZnS Qdot was used as the fluorophore reporter and AuNPs was used as the fluorescence quencher. The infectious state of individual host cell was quantitatively monitored both visually and via fluorescence intensity change in the concentration range of 1–200 plaque forming unit.