Standard and AEGIS nicking molecular beacons detect amplicons from the Middle East respiratory syndrome coronavirus.

This paper combines two advances to detect MERS-CoV, the causative agent of Middle East Respiratory Syndrome, that have emerged over the past few years from the new field of "synthetic biology". Both are based on an older concept, where molecular beacons are used as the downstream detection of viral RNA in biological mixtures followed by reverse transcription PCR amplification. The first advance exploits the artificially expanded genetic information systems (AEGIS). AEGIS adds nucleotides to the four found in standard DNA and RNA (xNA); AEGIS nucleotides pair orthogonally to the A:T and G:C pairs. Placing AEGIS components in the stems of molecular beacons is shown to lower noise by preventing unwanted stem invasion by adventitious natural xNA. This should improve the signal-to-noise ratio of molecular beacons operating in complex biological mixtures. The second advance introduces a nicking enzyme that allows a single target molecule to activate more than one beacon, allowing "signal amplification". Combining these technologies in primers with components of a self-avoiding molecular recognition system (SAMRS), we detect 50 copies of MERS-CoV RNA in a multiplexed respiratory virus panel by generating fluorescence signal visible to human eye and/or camera.

Introduction

MERS-CoV detection

The coronavirus that causes Middle East respiratory syndrome (MERS-CoV) is highly contagious, causing respiratory infection with high morbidity and mortality (Cheng et al., 2007, Chan et al., 2013). It first appeared in Middle East where recent reports suggest that camels are its most likely reservoir (Zaki et al., 2012, Chan et al., 2015). The most noted cases outside of that region have been in Korea, but several registered cases have appeared in the United States by “virus tourism”, including in Orlando, Florida (http://www.who.int/csr/don/archive/disease/coronavirus_infections/en/).

Outside the Middle East, MERS-CoV must be diagnosed by physicians who have very little experience with its symptoms. This creates a need for a rapid molecular test for the virus, to support the early diagnosis that is needed to arrange for quarantines, manage patient care, and help public health officials track disease outbreaks.

Several molecular tests for MERS-CoV detection have been created since the 2012 American outbreak (Corman et al., 2012a, Corman et al., 2012b, Lu et al., 2014). These generally use real-time RT-PCR amplification. For positive diagnosis of a MERS-CoV infection, amplification and detection of two different genomic targets is required, or sequencing that confirms a single target amplicon (http://www.who.int/csr/disease/coronavirus_infections/case_definition/en/). Culturing MERS-CoV from respiratory tract, blood, urine, or stool samples remains an option, although virus culturing is cumbersome, hazardous, and requires experienced staff to interpret cytopathic effects in host cells. Serological assays for detection of specific neutralizing anti-MERS-CoV antibodies are not ideal for early diagnosis and containment of the spread of the outbreaks. Therefore, RT-PCR based diagnostics still remains to be the most common detection method (Leland and Ginocchio, 2007, Chan et al., 2015).

Because of its instrument and power requirements, PCR cannot be used everywhere, including “low resource” environments. Therefore, tests that use isothermal amplification have been developed. These include reverse transcriptase loop-mediated isothermal amplification (LAMP) (Shirato et al., 2014) and reverse transcription recombinase polymerase amplification assays (RPA) (Abd El Wahed et al., 2013, Sharma et al., 2014). These require short incubation times, and are highly sensitive and specific, making them useful where no PCR equipment is available or operable.

Low resource environments also need inexpensive ways to detect the amplicons (Niemz et al., 2011, Boonham et al., 2014). Detection with intercalating dyes like SYBR green (Lu et al., 2014), measurements of turbidity generated by precipitating Mg2P2O7 during amplification (Fukuda et al., 2007, Tanner et al., 2012), or luciferase-mediated bioluminescence generated from the inorganic pyrophosphate (Gandelman et al., 2010) are cheap enough, but can be deceived by off-target amplicons. Thus, melting profiles are often needed to discriminate primer artifacts from amplicon, creating extra steps, extra complexity, and extra cost.

Therefore, molecular hybridization-based methods have been favored for amplicon identification. These have sequence-specificity, reducing the likelihood of false positives. In various architectures, oligonucleotide probes may be cleaved during the amplification reaction (e.g. TaqMan) (Schultz et al., 2011), or probes/primers that change conformation upon binding to the amplicons are exploited such as scorpion primers (Carters et al., 2008) or molecular beacons (Leone et al., 1998, Zou et al., 2013).

Molecular beacons (MB)

Molecular beacons are DNA molecules with a loop complementary to the DNA or RNA (collectively xNA) analyte and a fluorescent moiety at one end and a quencher moiety at the other. A hairpin stem brings the quencher close to the fluorophore, quenching its fluorescence. When the amplicon hybridizes to the loop, it opens the beacon, separating the fluorophore from the quencher, and increasing the overall fluorescent signal (Tyagi and Kramer, 1996, Tan et al., 2004).

Thus, the signal-to-noise ratio of beacon signaling is controlled by (a) the effectiveness of the quenching in the closed stem, (b) the extent to which the open and closed beacon are in equilibrium in the absence of analyte, (c) the possibility that non-target xNA might invade the stem and open it even in the absence of analyte, and (d) the amount of fluorescence when the analyte is bound. That ratio of fluorescence from an open/closed beacon is typically between 10 and 100 (Vet and Marras, 2005).

Two approaches have been introduced to improve this ratio. For example, to prevent invasion of the stem by non-target xNA, components of an artificially expanded genetic information system (AEGIS) have been placed in the stem (Sheng et al., 2008) (Fig. 1 A). AEGIS nucleotides rearrange the hydrogen bonding pattern of standard nucleotides, creating up to eight additional nucleotides that add up to four nucleobase pairs to the two pairs (A:T and G:C) found in natural nucleotides. AEGIS oligonucleotides cannot hybridize to any natural xNA, removing (c) as a factor in generating background noise. Further, AEGIS-AEGIS pairs are more stable than classical standard nucleobase pairs, minimizing background arising from (b) (Yang et al., 2006).

Fig. 1

(A) Components of AEGIS that, by rearranging hydrogen-bonding patterns on the nucleobases, adds six nucleotides to the four standard nucleotides. The Z:P pair was used in this assay. (B) Target DNA amplicon and complementary molecular beacon loop contains the recognition sequence of a nicking enzyme (Nt.BsmAI). When the target strand hybridizes to loop of the beacon, a full nicking enzyme recognition site forms where only MB is cut. After nicking enzyme cuts the beacon, hybrid strand becomes unstable and cleaved beacon dissociates from the target. Freed target strand can then be substrate for another molecular beacon to initiate the second cycle of cleavage thereby allowing exponential signal amplification. (C) Schematic showing that by strategic removal of hydrogen bonding groups, a SAMRS can be obtained to allow multiplexing.

The consequence of a more tightly bound stem is, however, that a longer loop is required for hybridization to pull the stem apart. This problem can be mitigated by introducing a site in the loop that, when hybridized to the analyte, becomes a substrate for a nicking enzyme that cuts the loop (Fig. 1B). In the absence of cut loop, the stem is much too short to hold the fluorophore and the quencher together, so they separate as independent molecules, generating the maximum fluorescence possible.

As a further advantage to the nickase architecture, after nicking, the analyte also dissociates from the cut loop, now free to bind a second beacon loop, anneal, and direct the nicking enzyme to cleave a second beacon. This allows a single target analyte to activate more than one beacon, providing signal amplification, often by a factor of 10 (Zheleznaya et al., 2006, Li et al., 2008, Connolly and Trau, 2010).

Here we report the combination of these strategies to improve signal-to-noise with molecular beacons. The beacons contained both AEGIS in the stem and a target-specific site recognized by a nicking endonuclease in the beacon loop. To detect MERS-CoV in a respiratory panel targeting Influenza A (InfA), Influenza B (InfB), Human respiratory syncytial virus (RSV), Severe acute respiratory syndrome coronavirus (SARS-CoV) and MERS-CoV, we included SAMRS nucleotides (Fig. 1C) in the up-front PCR’s primers (Hoshika et al., 2010). SAMRS nucleotides are designed not to interact with each other. Thus, SAMRS oligonucleotides added to a multiplexed nucleic acid targeted assays cannot create primer-dimer artifacts. Combined with AEGIS technology, SAMRS gives cleaner and more robust responses in assays (Yang et al., 2013, Glushakova et al., 2015a, Glushakova et al., 2015b). Here, the combination of three innovations allowed us to successfully detect 50 viral RNA copies of MERS-CoV in authentic samples obtained from NIAID in Frederick, MD.

Materials and methods

Design and synthesis of oligonucleotide primers and molecular beacons

Primers and molecular beacons were designed with StrainTargeter, an in-house software package (Table 1, Table 2 ). StrainTargeter analyzes multiple sequence alignments (MSAs) of virus families built from public databases (GenBank, http://www.ncbi.nlm.nih.gov/genbank; ViPR, http://www.viprbrc.org; and FluDB, http://www.fludb.org), finding regions within those viral genomes that have a level of sequence divergence that allows viral targets to be distinguished, but not so much to prevent detecting viruses that are divergently evolving. A BLAST search then follows to ensure that primer and probe sequences designed by StrainTargeter are not closely similar to sequences in both the NCBI RNA virus database and the NCBI human genome database.

Table 1

Hybrid SAMRS-AEGIS primers and MERS-CoV amplicons for respiratory virus panel.

Virus	Sequences (5′-3′)	Targeted region
Human respiratory syncytial virus (RSV)	Forward primer: CTAPTCCPCCAPCPAPCGGGCAAATATGGAAACATA*C*G*T*GReverse primer:CAGPAAGPGGTPGPTPG GGAACATGGGCACCCAT*A*T*T*G	Pneumovirus matrix protein gene
Severe acute respiratory syndrome coronavirus (SARS-CoV)	Forward primer:CTAPTCCPCCAPCPAPC GAGGAGGTTGTTCTCAAG*A*A*C*GReverse primer:CAGPAAGPGGTPGPTPG GTAAACCAGGAGACAAT*G*C*G*C	Orf1ab polyprotein gene
Influenza A (InfA)	Forward primer:CTAPTCCPCCAPCPAPCCATGGAATGGCTAAAGACAA*G*A*C*CReverse primer:CAGPAAGPGGTPGPTPG CAAAGCGTCTACGCT*G*C*A*G	Segment 7 matrix protein 2 (M2) and matrix protein 1 (M1) genes
Influenza B (InfB)	Forward primer:CTAPTCCPCCAPCPAPC GATGGCCATCGGATCC*T*C*A*AReverse primer:CAGPAAGPGGTPGPTPGTAATCGGTGCTCTTGACCAA*A*T*T*G	Segment 8 nuclear export protein (NEP) and nonstructural protein 1 (NS1)
Middle East respiratory syndrome coronavirus (MERS-CoV) 1-2	Forward primer:CTAPTCCPCCAPCPAPCCATGCTATTGCTTTAACGCTG*A*G*G*GReverse primer:CAGPAAGPGGTPGPTPGGTCTCAGAAATGCACTCTGATTCAC*C*T*T*C	ORF1a
MERS-CoV 6-1	Forward primer:CTAPTCCPCCAPCPAPC CTGGCATTGTAGCAGCTGTT*T*C*A*GReverse primer:CAGPAAGPGGTPGPTPG GAGTGGACGTACGACAGTTG*T*A*C*C	N and ORF8b
Amplicon MERS-CoV 1-2 (145 bp)	CATGCTATTGCTTTAACGCTGAGGGTGATGCATCTTGGTCTTCTACTATGATCTTCTCTCTTCACCCCGTCGAGTGTGACGAGGAGTGTTCTGAAGTAGAGGCTTCAGATTTAGAAGAAGGTGAATCAGAGTGCATTTCTGAGAC
Amplicon MERS-CoV 6-1 (159 bp)	CTGGCATTGTAGCAGCTGTTTCAGCTATGATGTGGATTTCCTACTTTGTGCAGAGTATCCGGCTGTTTATGAGAACTGGATCATGGTGGTCATTCAATCCTGAGACTAATTGCCTTTTGAACGTTCCATTTGGTGGTACAACTGTCGTACGTCCACTC

Note: A*, T*, G*, C*: SAMRS nucleotides (Fig. 1B) and P: AEGIS nucleotide (Fig. 1C). MB recognition site is in bold.

Table 2

Molecular Beacons and probes.

Name	Targeted region	Sequence (5′-3′)
Std MB-MERS6-1	MERS 6-1	FAM-CGTGCGCAATTAGTCTCA^GGATTGAATGACCA CGCACG-BHQ1
AEGIS MB-2P5Z_MERS6-1	MERS 6-1	FAM-CPTGZG CAATTAGTCTCA^GGATTGAATGACCA CPCAZG-dabcyl
Target probe	MB loop	TGGTCATTCAATCCTGAGACTAATTG
MBstemtarget-6	Stem	CGTGCG
MBstemtarget-8	Stem + partial loop	CGTGCGTG
MBstemtarget-10	Stem + partial loop	CGTGCGTGGT
MBstemtarget-12	Stem + partial loop	CGTGCGTGGTCA

Note: Nicking enzyme recognition site in bold, nicking site is marked as . MB stem sequence is underlined. P and Z: AEGIS nucleotides (Fig. 1C).

Two nicking molecular beacons (standard and AEGIS) were designed to detect MERS-CoV, with 6 nucleotide long stem and 26 nucleotides long loop sequence where nicking site (Nt.BsmAI, NEB) was centralized. For the AEGIS MBs, dC: dG pair in its stem was replaced by dZ: dP pair. The corresponding standard RP-HPLC purified molecular beacon purchased from Integrated DNA Technologies (IDT, Coralville, IA). It was labeled with a fluorophore, FAM, at the 5′-end, and a quencher, Black Hole Quencher-1 (BHQ-1), at 3′-end. The AEGIS molecular beacon labeled with a fluorophore, FAM, at the 5′-end, and a quencher, DABCYL, at 3′-end, was synthesized in house on ABI 394 synthesizer and RP-HPLC purified (Table 2).

Detection limit of molecular beacons

Detection sensitivity of each molecular beacon was first determined by hybridization of a short 26mer target probe (IDT, Coralville, IA) complementary to loop of the two beacons. Beacons (500 nM) were mixed with varying concentrations of complementary probe (500 nM, 200 nM, 100 nM, 50 nM, 25 nM, 10 nM, 5 nM, 1 nM, 0.5 nM, 0.1 nM and 0.05 nM) in hybridization buffer (50 μL, CutSmart® buffer, NEB: 50 mM KOAc, 20 mM Tris-acetate, 10 mM MgOAc, 100 μg/mL BSA, pH 7.9, 25 °C). Thermal profiles were recorded as follows: Mixtures were first denatured at 95 °C for 1 min. The temperature was then decreased to 85 °C and held for 0.5 min. Then, the temperature was decreased slowly from 85 °C to 25 °C (programmed to occur over 70 min), with fluorescence continuously measured using Roche Light Cycler 480. For nicking enzyme-assisted signal amplification (NESA), Nt.BsmAI (20 units, 5 U/μL) was added and samples were incubated at 37 °C for about 30 min, then at 55 °C for 90 min with continuous fluorescence monitoring. Additionally, images of fluorescence generated by the beacons induced by blue LED light (470 nm) at room temperature were recorded through an orange filter by a digital camera.

MERS-CoV RNA isolation

MERS-CoV RNA was isolated according to a standard protocol (Life Technologies). Briefly, Trizol-inactivated MERS-CoV (Jordanian isolate, GenBank accession no. KC164505.2) was obtained from Lisa Hensley and Reed Johnson of the National Institute of Allergy and Infectious Diseases (Fort Detrick, Frederick, MD, USA) at a titer of 2 × 104  pfu/mL. First, chloroform (0.2 mL) was added to the virus homogenate in Trizol (1 mL). The tube was briefly vortexed (15 s) and incubated (∼3 min, at room temperature). The sample was centrifuged at 12,000g for 15 min at 4 °C, aqueous phase collected, and glycogen added (5 mg/mL). Viral RNA was then precipitated by adding an equal volume of isopropanol at room temperature for 10 min. The RNA was pelleted by centrifugation at 10,000g for 10 min at 4 °C, the supernatant was removed, and the pellet was washed twice with 75% ethanol. Finally, the pellet was dissolved in the nuclease-free water (Life Technologies), aliquoted, and stored at −80 °C.

Reverse transcription PCR

Mono- (MERS-CoV specific primers only) or multiplexed one-step RT-PCR reactions (combination of MERS-CoV and RSV, SARS-CoV, InfA, InfB primers, (Table 1) including MERS-CoV RNA (100 copies) were carried out by SuperScript One-Step RT-PCR kit with Platinum Taq (Life Technologies, Carlsbad, CA). Reactions took place in 1 x reaction mix with additional MgSO4 (1.5 mM, final volume, 20 μL). The reaction mixture contained 0.3 μM of each sets of forward and reverse hybrid SAMRS-AEGIS target-specific primers, and 2.5 U of RT/Platinum Taq enzyme mix. RT-PCR cycling conditions were as follows: 1 cycle of the complementary DNA (cDNA) synthesis and pre-denaturation (55 °C for 30 min and 94 °C for 2 min), 35 cycles of PCR (94 °C for 15 s, 56 °C for 30 s, and 70 °C for 30 s), and final extension at 72 °C (5 min).

Forward or reverse primer extension reactions

To favor the generation of single stranded amplicons prior to molecular beacon detection, a primer extension reaction was performed using only forward or reverse primers. Molecular beacons were designed to target the amplicons generated by forward primer (sense strand) whereas amplicons generated by reverse primer (antisense strand) served as a negative control in MB detection experiments. RT-PCR product (20 μL) was directly added to the primer extension reaction. Primer extension reaction (200 μL) was carried in 1 x ThermoPol Buffer (20 mM Tris-HCl, 10 mM (NH4)2SO4, 10 mM KCl, 2 mM MgSO4, and 0.1% Triton X-100, pH 8.8) at 25 °C (NEB, Ipswich, MA) with forward or reverse hybrid AEGIS-SAMRS primer (0.2 μM), dNTPs (0.2 mM each) and Vent (exo-) DNA polymerase (1 U per reaction; NEB). Cycling conditions were as follows: 95 °C (1 min), followed by 20 cycles (94 °C for 20 s, 56 °C for 30 s, and 72 °C for 30 s) with a final incubation cycle at 72 °C for 1 min. Reaction mixtures were then quenched with 4 mM EDTA. Amplicons (5 μL) were run on 2.5% TBE-agarose gel for visualization.

Molecular beacon detection of MERS-CoV amplicons

Primer extension products were purified by using DNA clean and concentrator kit (Zymo Research, Irvine, CA) and eluted in 50 μL of nuclease–free water. Eluted samples were divided into half (25 μL) and then mixed with 500 nM MBs (Std or AEGIS) in CutSmart® hybridization buffer (50 μL, NEB). Thermal profiles were first recorded the same way as for initial MB hybridization studies. For NESA, 20 units of Nt.BsmAI (5 U/μL, NEB) was added and samples were incubated at 37 °C for about 90 min, then at 55 °C until fluorescence curve reached saturation, with continuous fluorescence measurement. Additionally, images of fluorescence generated by a molecular beacon (with and without nicking enzyme, LED at 470 nm) at room temperature and were recorded through an orange filter by a camera.

Stem invasion of molecular beacons

To show that MB stems containing dZ and dP resist stem invasion and thereby reduce background noise, MBs (500 nM) were incubated with 2-fold excess of synthetic oligonucleotides of various lengths (IDT, 6mer, 8mer, 10mer, 12mer), each having six nucleotides complementary to the stem of the MB and a remaining sequence matching the loop region (Table 2). Thermal profiles were recorded after separately incubating these with both standard and AEGIS MBs.

Results

Concept of nicking enzyme signal amplification

For target-specific DNA/RNA detection, we used a special family of restriction endonucleases, DNA nicking enzymes. Nicking enzymes function as restriction enzymes, except that they cut only one strand of the recognition duplex instead of both strands (Fig. 1B).

Here we exploited a segment of the MERS-CoV virus that had the recognition sequence of Nt.BsmAI nicking enzyme. The loop of the molecular beacon contains its complement with the cleaved site. Thus, when the amplicon generated by primer extension PCR hybridizes to the loop portion of the beacon, a full nicking recognition site forms where the beacon loop (not the amplicon) is cut. After the nicking reaction, the freed target strand is then available to hybridize to the loop of another molecular beacon, causing multiple cleavages, resulting in signal amplification.

Evaluation of standard and AEGIS molecular beacons

We first examined the sensitivity of the beacons in the absence of nicking enzyme. Thermal profiles were recorded by incubating standard or AEGIS MBs (500 nM) with 26mer target probe at variable concentrations ranging from 500 nM to 0.05 nM (Table 2). In the absence of the nicking enzyme, standard MB can only detect down to 200 nM (6 × 1012 copies) of target probe (Fig. 2 A and Fig. 2C) while the AEGIS MB was more sensitive and was capable of detecting 100 nM target concentration (3 × 1012 copies) as have been judged by the fluorescence visible to human eye. At the lower target concentrations no significant increase in fluorescence intensity was observed.

Fig. 2

(A) and (C) Thermal profiles of standard or AEGIS MBs were recorded with varying concentrations (0 to 500 nM) of short target complementary to MB loop. Time courses for NESA were recorded when nicking enzyme (Nt.BsmAI) and standard (B) or AEGIS MBs (D) were incubated with varying concentrations of short target DNA. Fluorescence intensity of each sample (before and after NESA) is recorded through an orange filter by a camera at room temperature.

We then repeated the assay using Nt.BsmAI. As Nt.BsmAI is optimally active at 37 °C, the mixture was first incubated at 37 °C (90 min). Then, the temperature was increased to 55 °C to help dissociation of cleaved components of molecular beacon and to favor further increase in fluorescence intensity.

Use of Nt.BsmAI improved the limit of detection (LOD) to 5 nM (1011 copies of target) for both MBs (Fig. 2  B and Fig. 2D). Fluorescence continued to increase up to a plateau over ∼1 h, indicating beacon cleavage turnover. With complete nicking, fluorescence had increased ∼14-fold. Thus, each target molecule evidently cleaved ∼100 beacons, as the target:beacon ratio was 1:100 (5 nM to 500 nM). This result shows that NESA assay enables detection of otherwise undetectable amounts of target.

MERS-CoV detection with molecular beacons

To detect the MERS-CoV in real samples, we extracted primers previously used for a RT-PCR based respiratory panel targeting InfA, InfB, RSV, SARS-CoV and MERS-CoV (Glushakova et al., 2015b). These SAMRS-AEGIS forward and reverse primer pairs were tested by single-plexed assays (MERS 6-1), 2-plexed (MERS 6-1 and MERS 1-2), and 6-plexed assays (MERS-CoV 6-1, MERS-CoV 1-2, RSV, SARS-CoV, InfA and InfB) in one-step RT-PCR with 100 copies of full genomic MERS-CoV RNA followed by forward or reverse primer extension to generate single sense and antisense strands, respectively (Table 1). Each extension reaction produced the amplicon of the expected size, which was confirmed by agarose gel electrophoresis followed by ethidium bromide staining (Fig. 3 A).

Fig. 3

(A) The sizes of single-plexed or multiplexed RT-PCR generated amplicons were confirmed by ethidium bromide staining followed by 2.5% TBE-agarose gel electrophoresis. (B) and (D) Thermal profiles of standard and AEGIS MBs were recorded with amplicons from forward or reverse primer extension reactions, respectively. (C) and (E) Time courses for NESA were recorded when nicking enzyme Nt.BsmAI and amplicons were incubated with standard or AEGIS MBs, respectively. Fluorescence intensity of each sample (before and after NESA) was recorded through an orange filter by a camera. NTC is MB only in hybridization buffer.

The identity of each amplicon was then confirmed by the hybridization with the target specific molecular beacons followed by NESA. Since we tested standard and AEGIS molecular beacons simultaneously, each beacon detected half of the starting MERS-CoV viral load, 50 copies in each case. Molecular beacons with the nicking sites embedded in their loops targeted the amplicon strand generated from the forward primer (“sense”). The “antisense” amplicon served as a negative control. As seen in Fig. 3B and D, both beacons successfully detect “sense” amplicons in 1-plexed, 2-plexed and 6-plexed assays.

Signal amplification was more pronounced when nicking enzyme was included in the detection assay. Moreover, this assay was target-specific, because amplicons from the opposite strand failed to produce a change in fluorescence (Fig. 3C and E). It is also notable that total fluorescence intensity was higher by 1.6-fold when the MERS-CoV amplicon was used as a target rather than a synthetic RNA simulant. This showed that in addition to being complementary to the beacon loop, the amplicon from the MERS-CoV genome had an additional nucleotide complementary to the first (standard) nucleotide in the stem (Figs. 2 D and 3 E). This improved the efficiency of target hybridization, NESA-induced cleavage, and downstream dissociation of the cleaved fragments.

Most importantly, the AEGIS MB exhibited higher fluorescence signal than the standard MB. Total fluorescence intensity increased by 1.8-fold when MERS-CoV viral genome was present (Fig. 3C and E).

Stem invasion assay

To show that MB stems with dZ and dP reduce the background fluorescence by resisting the stem invasion, MBs (500 nM) were incubated with 2-fold excess of synthetic oligonucleotides having various lengths (6mer, 8mer, 10mer, 12mer) that had six nucleotides complementary to the stem of the MB, and a remaining sequence matched the loop region (Table 2). Thermal profiles were recorded after separately incubating these with both standard and AEGIS MBs and camera images are also taken at room temperature for visualization. For standard MB, incubation with 2-fold excess of the 6mer failed to open the beacon, while 8mer DNA target exhibited slight increase in fluorescence. This was consistent with the design strategy that intramolecular binding constant between the stem sequences is far greater than the intermolecular interaction between one arm of the stem and short DNA targets (6mer or 8mer). On the other hand, in the presence of the 10mer and 12mer targets, stem and part of the loop of the MB hybridized to the target, disrupted the hairpin structure, and increased the background fluorescence by 24% and 62%, respectively (Fig. 4 A). In contrast, AEGIS-containing MB was not opened by any of the stem targets tested except with the 12mer target where only slight increase in fluorescence is observed (Fig. 4B). These results suggest that AEGIS-modified molecular beacons are superior to standard beacons in terms of resistance to stem invasion and overall NESA efficiency.

Fig. 4

Stem invasion experiments. Thermal profiles of (A) standard and (B) AEGIS MBs incubated with synthetic oligonucleotides in various lengths that had six nucleotides complementary to the stem of the MB, and a remaining sequence matched the loop region, were recorded followed by a digital camera imaging.

Discussion

Multiplexed PCR-based assays that target the multiple possible causes of respiratory infection often fail because multiple primers, presented at high concentrations, interact with each other to divert amplification resources and form artifacts, including primer-dimers and extension products of nonspecifically paired primers, often imperfectly matched. These non-specific reactions often lead to false negatives in nucleic acid-targeted assays (Elnifro et al., 2000).

To support the high levels of multiplexing, we previously developed SAMRS for primers (Hoshika et al., 2010). Here we showed again that SAMRS-AEGIS primers improved the performance of the multiplexed RNA-targeted amplification.

In addition to SAMRS-AEGIS technologies, we explored a new signal amplification mechanism to improve the sensitivity of the molecular beacons. Molecular beacons have an advantage over traditional DNA probes by generating an “on” or “off” signal. Ideally, they do not emit fluorescence until target is hybridized to beacon to separate quencher from the fluorophore. Although their simplicity of function makes them attractive for direct detection, their intrinsic background and interaction with adventitious xNA limits their sensitivity to high nano-molar concentrations of target (Zhang et al., 2001, Drake and Tan, 2004).

Thus, their ability to detect low concentrations of DNA or RNA benefits from amplification at the detection step as well. This was obtained here using a nicking enzyme, boosting the sensitivity by 1–2 orders of magnitude, enough to allow a signal to be visible from just 1010–1011 copies of amplicon.

The sensitivity of NESA was further improved, especially in complex assay mixtures, by adding AEGIS nucleotides to the beacon stems. This eliminated unwanted stem invasion by DNA complementary to the nucleotide sequence in the stem.

In a combination using AEGIS in a nicking molecular beacon, we were able to detect 50 copies of MERS-CoV in a real sample, and in an assay that could also detect influenza, RSV and other less exotic species that might cause the respiratory infection in an admitted patient. This allows a physician to rapidly distinguish a patient having one of these common infections from a patient that has an unusual (at least in the United States) MERS-CoV infection.