Recognition of Dual Targets by a Molecular Beacon-Based Sensor

DOI: 10.1021/acs.analchem.5b00810. Publication Date (Web): April 16, 2015. Copyright © 2015 American Chemical Society. *Phone: +886-4-2359-0121, ext...
0 downloads 5 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Recognition of Dual Targets by a Molecular Beacon-Based Sensor: Subtyping of Influenza A Virus Chun-Ching Lee,† Yu-Chieh Liao,‡ Yu-Hsuan Lai,† and Min-Chieh Chuang*,† †

Department of Chemistry, Tunghai University, Taichung 40704, Taiwan Institute of Population Health Science, National Health Research Institutes, Zhunan, Miaoli County 35053, Taiwan

‡

S Supporting Information *

ABSTRACT: A molecular beacon (MB)-based sensor to offer a decisive answer in combination with information originated from dual-target inputs is designed. The system harnesses an assistant strand and thermodynamically favored designation of unpaired nucleotides (UNs) to process the binary targets in “AND-gate” format and report fluorescence in “off−on” mechanism via a formation of a DNA four-way junction (4WJ). By manipulating composition of the UNs, the dynamic fluorescence difference between the binary targets-coexisting circumstance and any other scenario was maximized. Characteristic equilibrium constant (K), change of entropy (ΔS), and association rate constant (k) between the association (“on”) and dissociation (“off”) states of the 4WJ were evaluated to understand unfolding behavior of MB in connection to its sensing capability. Favorable MB and UNs were furthermore designed toward analysis of genuine genetic sequences of hemagglutinin (HA) and neuraminidase (NA) in an influenza A H5N2 isolate. The MB-based sensor was demonstrated to yield a linear calibration range from 1.2 to 240 nM and detection limit of 120 pM. Furthermore, high-fidelity subtyping of influenza virus was implemented in a sample of unpurified amplicons. The strategy opens an alternative avenue of MB-based sensors for dual targets toward applications in clinical diagnosis.

W

absence of target DNA, leading to close proximity between the fluorophore and the quencher (fluorescence quenching). Upon hybridization with a complementary target DNA in the loop region, the stem duplex unwinds to restore the fluorescence emission. Thus, a sensor with “off−on” system is designed. In the multiplexed assays, numerous MBs characteristic of different targets with varied fluorescent tags are required (Scheme 1A). This can be problematic as a result of limited options of fluorophores, the overlap of fluorescence spectra, and the restrictions of available optical channels in current platforms.8,15,16 To the necessity of multi-input analysis, a novel concept of molecular logic computation has emerged as a versatile solution.17 Yet there have been very limited studies addressing multisequence identification with intelligent configuration which can suffice for the need in biomedical applications.11,18−20 The present work reveals a strategy based on a single MB, in cooperative operation with an assistant strand (A) and preferable combination of unpaired nucleotides (UNs), to logically process dual inputs of biological significance (Scheme 1B). The assistant strand contains two fragments (A-I and A-II)

ith the increasing understanding of nucleic acid sequences and functions, it is recognized that combining information from multiple genetic fragments is beneficial to medical interpretations and decisions. One of the notable examples is the expression of miRNAs, which has been discovered to have profound effects on cellular differentiation, biological functions, and cancer development, in a largely combinatorial manner.1−3 Another example of clinical significance is the drug resistance in Mycobacterium tuberculosis (Mtb) infection. It has been reported that there are numerous mutations (or minor insertion) in the rpoB gene of Mtb responsible for the resistance against a specific antibiotic treatment.4−6 To satisfy the demands for analysis of multiple nucleic acids, myriads of multiplexed assays have been reported, most of which are based on the real-time polymerase chain reaction (rtPCR). Specifically, in most multiplexed methods, multiple signals are outputted in response to the input sequences in a “single input, single output” fashion.7−10 However, this fashion of signal derivation perplexes the information integration and is often uneconomic and redundant to answer a single question in a practical manner.11 Molecular beacon (MB), a stem−loop-structured DNA oligonucleotide modified with a fluorophore and a quencher at its two ends,12−14 has been a reporter probe extensively incorporated in rtPCR. MB folds in a hairpin conformation in © XXXX American Chemical Society

Received: March 2, 2015 Accepted: April 16, 2015

A

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

■

EXPERIMENTAL SECTION Synthesis and Design of Oligonucleotides and MBs. The molecular beacons 1 and 2 (MB1, MB2) were synthesized by Integrated DNA Technology (Coralville, IA). Both MBs featured a 6-carboxyfluorescein (6-FAM) attached to the 5′ end and a 4-(4′-dimethylaminophenylazo)benzoic acid (Dabcyl) quencher attached to the 3′ end. Other oligonucleotides were obtained from MDBio Inc. (Taipei, Taiwan) and purified by HPLC reversed-phase cartridge. The sequences of all oligonucleotides are provided in Supporting Information Table S-1. Thermal Denaturation Profile of MBs. Thermal denaturation profiles of MB1 and MB2 were obtained using the MiniOpticon Real-Time PCR system (Bio-Rad, Hercules, CA). Excitation wavelength was fixed at 470−500 nm to measure the emitted fluorescence at wavelengths of 523−543 nm. The fluorescent signal changes of the MBs alone or in incubation with the target hybrids were examined by mixing 0.2 μM MB with desired targets (1.2 μM) or/and assistant probes (1.2 μM) in a buffer composed of 10 mM Tris, 50 mM NaCl, and 5 mM MgCl2. The temperature was commenced at 95 °C for 3 min, and then increased from 15 to 85 °C at 1 °C increment. Each temperature was lasted for 5 min. Equilibrium Analysis. Equilibrium constants for the hybridization between the MB and its target hybrids were evaluated based on the method described previously.22,23 Briefly, in a solution containing MB and target DNA (T), there existed at least three distinct states: MB-T hybrid (phase 1, denoted as MB·T), stem−loop folded MB free from the target (phase 2, MBclosed + T), and random coiled MB free from the target (phase 3, MBopen + T). The reaction can be presented as

Scheme 1. (A) Schematic Representation of the Three Phases in the Solution Consisting of Conventionally Designed Molecular Beacons (MBs) and Complementary Target DNA in “Single Input, Single Output” Fashion (Phase 1, MBs−Target Hybrid; Phase 2, Closed MBs Dissociated from Targets; Phase 3, Random-Coiled MBs Free from Targets; Red, BlackMBs; Green, BlueTarget Sequences); (B) Phase Transitions of the Assistant StrandCooperative Association among the Molecular Beacon (Red), the Assistant Strand (Pink), and the Dual-Target Sequences (Blue and Green) (Phase 1, Formation of the Four-Way Junction (4WJ); Phase 2, Closed MB Dissociated from Assistant Strand/Targets Hybrid; Phase 3, RandomCoiled MB, Assistant Strand, and Targets; the Assistant Strand Contains Two Fragments (A-I and A-II) Complementary to Hanging Segments (III and IV) of the Dual-Target DNAs)

K12

K 23

MB·T HooI MBclosed + T HooI MBopen + T

(1)

K12 and K23 can be expressed as

K 23 = K12 =

complementary to hanging segments (III and IV) of the target DNA to enable formation of a DNA four-way junction (4WJ) with MB (phase 1 of Scheme 1B). Furthermore, UNs are left on both targets residing in the center of the 4WJ to confer structural flexibility facilitating formation of this conformation. Then, MB unwinds and emits fluorescence only in the concurrent presence of the dual targets. Influenza, an acute respiratory and highly contagious disease, has caused substantial morbidity and mortality globally at least since the Middle Ages.21 Particularly identifying the strain which is either highly pathogenic or leading to high mortality necessitates the simultaneous recognition of hemagglutinin (HA) and neuraminidase (NA). Accordingly, influenza A virus strain [A/duck/Taiwan/DV30-2/2005 (H5N2)] was utilized as a paradigm to interrogate the sensing capability of the present system for dual targets. To our best knowledge, this is the first work harnessing a single MB to simultaneously recognize dualtarget sequences with biomedical significance. The strategy was demonstrated to successfully implement subtyping of influenza viruses in samples of unpurified amplicon.

[MBopen] [MBclosed]

(2)

[MBclosed][T] [MBT]

(3)

and derived to be represented in terms of fluorescence (see Supporting Information for details). Thermodynamic Analysis. The thermodynamic parameters (ΔH12 and ΔS12) that describe the transition from a molecular beacon−targets complex phase to a phase composed of free molecular beacons and target hybrids were determined by analyzing the fluorescence data obtained from the thermal denaturation profiles of 0.2 μM of molecular beacons incubated with 1.2, 1.8, 3, 4.8, and 7.2 μM of targets and assistant probes. Since ΔG = −RT ln(K12) = ΔH − TΔS and K12 = [T] while [MB] = [MB·T] = 1/2[MB0], the thermal denaturation data could be fitted to a straight line given by the equation R ln([T0] − 0.5[MB0]) = −

1 ΔH12 + ΔS12 θm

(4)

where the melting temperature θm is at which its characteristic K12 equals to the theoretical K12 calculated from K12 = [T]. On the basis of eq 4 displayed above, the slope of the fitted straight line represents the enthalpy change ΔH12 and the y-intercept represents the entropy change ΔS12. B

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 1. (A) Thermal denaturation profiles of the MB1 solution (black) containing target TG (wine), target TI (orange), TG and TI (magenta), assistant strand A9 and TG (blue), A9 and TI (olive), as well as A9, TG, and TI (red). Inset: magnified plots for the curves given at temperatures from 30 to 55 °C. (B) Dynamic curves (32 °C) of the MB1 unfolding in the absence (black) and presence of the assistant strand, TG, and TI, with varied combinations of unpaired nucleotides (UNs): red (1,1), blue (2,0), olive (2,1), and magenta (0,0). The numbers in the parentheses denote the number of UNs left in TG and TI. The arrow indicates the time point when the assistant strand/targets hybrid was injected. The concentration of MB1 was 0.2 μM. The concentration for assistant probes, TG, and TI was 1.2 μM. (C) Schematic configuration of the four-way junction (4WJ) composed of molecular beacon 1 (MB1), the assistant strand (A9), and the targets TG and TI. The target fragments hybridizing with loop are highlighted with red (TG) and blue (TI). The assistant strand is represented with green. Inset: envisaged diagrams depicting the sequence composition in the center of the 4WJ with varied unpaired nucleotides combinations.

μM MB and 1.2 μM the assistant strand in a buffer composed of 10 mM Tris, 50 mM NaCl, and 5 mM MgCl2. Temperature was elevated to 95 °C for 3 min and subsequently increased from 42 to 50 °C at a 0.2 °C/min increment rate. Fluorescence intensity recorded at 45 °C was regarded as the output signal characteristic of the target concentration. Analysis of aPCR Products. The analysis was conducted using the MiniOpticon Real-Time PCR system. Excitation wavelength was fixed in 470−500 nm to measure the fluorescence of 523−543 nm. Aliquots of the aPCR products from H5 (8 μL) and/or N2 (8 μL) were mixed with 0.2 μM MB (2 μL) and 1.2 μM assistant probes (2 μL). In the cases given with either H5 or N2 alone, the volume was added up to 20 μL with a buffer composed of 10 mM Tris, 50 mM NaCl, and 5 mM MgCl2. The mixture was heated to 95 °C for 3 min and subsequently increased from 40 to 55 °C at a 0.2 °C/min increment rate. Fluorescence intensity recorded at 50 °C was regarded as the output signal for representation in histograms.

Kinetic Analysis. Dynamic curves to determine unzipping rate constants of MB with target sequences were obtained using FlexStation 3 multimode microplate reader (Molecular Device, Sunnyvale, CA). After the fluorescence measurement of the MB solution (0.33 μM, 60 μL in a 96-well plate) for 5 min, the FlexStation 3 system injected 40 μL of the target hybrid solution, which contained 3 μM of dual targets and the assistant probe and had been preliminarily incubated for 30 min, into the MB solution and pipetted three times successively. The experiments were performed at 32 °C. During the injection and pipetting, fluorescence recording was ongoing. The excitation and emission wavelengths were set at 490 and 525 nm, respectively, with a cutoff at 515 nm. To analyze the unfolding kinetics of MB upon hybridization with target DNAs, a second-order reaction was assumed in line with the prior work reported by Tsourkas et al.:23 MB + T ⇌ MB·T

(5)

d[MB·T] = k1[MB][T] − k 2[MB·T] dt

(6)

■

RESULTS AND DISCUSSION “Off−On” Mechanism against Binary Targets. As the goal of the present study aims an identification of specified influenza virus, recognition of both HA and NA genetic inputs is demanded and only positively outputted in their concurrent presence (an AND-gate fashion). With a cooperative assistant strand (A9) hybridizing with dual targets (TG and TI), molecular beacon 1 (MB1, see Supporting Information Table S-1 for detailed sequence) exhibits significant fluorescence (at 40−50 °C) only in the simultaneous presence of TG and TI (red curve) and gives minimal fluorescence intensity (over 15− 55 °C) in the other scenarios (Figure 1A). In the concurrent presence of TG, TI, and A9 (in a configuration as shown in Figure 1C), the fluorescence increase was trivial at 20−30 °C, indicating that MB1 was stably folded at low temperatures. Subsequently, a sharp rise at 30−47 °C was observed, which can be resulted from gradually declined stability of the stem region of MB1, in connection to elevated hybridization efficiency between the loop region and the TG/TI/A9 hybrid. The fluorescence declined remarkably with the rising temperature from 47 to 56 °C as a consequence of disruption of the 4WJ. Finally, the fluorescence restored at the temperatures higher than 56 °C due to the thermal denaturation of MB1. Different from the thermal denaturation curve illustrated

where k1 is the opening rate constant and k2 is the closing rate constant of MB upon hybridization. The k1 could be eventually derived by fitting a normalized fluorescence (see Supporting Information for details) and considered as unfolding rate constant (k) illustrated in the Results and Discussion section. Polyacrylamide Gel Electrophoresis (PAGE). The product collected from the incubation of the dual-target sequences and the assistant strand was analyzed to ascertain that the three-component hybrid is stably formed. Briefly, an aliquot comprising 0.2 μM of each target strand and assistant probe (in a buffer containing 10 mM Tris, 50 mM NaCl, and 5 mM MgCl2) was mixed thoroughly. This solution was then heated at 95 °C for 3 min and incubated at 32 °C for 30 min. The resulted products were examined by PAGE to confirm the presence of hybrids. DNA 4WJ and products of asymmetric polymerase chain reaction (aPCR) were also verified by PAGE following same protocols above. Detection of Targets with Varied Concentration. Determination of concentration was conducted using the MiniOpticon Real-Time PCR system. Excitation wavelength was fixed in 470−500 nm to measure the fluorescence of 523− 543 nm. Desired concentrations of targets were mixed with 0.2 C

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry previously22 in which substantial fluorescence was observed at 20−40 °C, MB 1 herein was designed in a relatively steady/ secure hairpin DNA configuration (ΔG = −3.57 kcal/mol at 32 °C, 50 mM Na+, and 5 mM Mg2+ predicted by mfold Web server) aiming an unfolding induced by the simultaneous presence of multiple target sequences in which undesired unzipping by a single target should be minimized. Taking the substantially great fluorescence given at ca. 47 °C into account, the system offers excellent “turn-on” sensing capability and responds only in the simultaneous presence of dual targets. Note that MB1 was also employed to study the hybridization with binary targets without assistant strand (Supporting Information Figure S-1). Slight fluorescence intensity (0.096 at 41 °C) was observed in the TG/TI combination (curve b, Supporting Information Figure S-1). Another two targets (TE and TF) with further extension (3-nt) hybridizing with the stem region were also applied (curve c, Supporting Information Figure S-1) to yield significantly higher fluorescence (10−50 °C). Nevertheless, either TE or TF alone caused substantial unfolding (inset of Supporting Information Figure S-1), which was contrary to the aim in establishing a dual-target analysis processing as an AND gate. This validates the utility of the assistant strand to realize the AND-gate operation. Optimization of Unpaired Nucleotides (UNs). Composition of UNs was found to be critical to MB unfolding. As UNs in the center serve as the cushioning junction for bending a long duplex, the number of UNs, in principle, determines the flexibility of the 4WJ and probability of its formation. A thermal denaturation profile given no UNs [denoted as (0,0) to represent the number of UNs left in TG and TI] is shown in Figure 2A. The maximum fluorescence (obtained from TG/TI/ A) occurred at approximate 51 °C; nevertheless, its intensity is only 40% of that given by (1,2) (Figure 1A). Without UNs, the tension in the junction was too high to form the 4WJ favorably.

The TG/TI/A fluorescence output in (1,1) or (2,0) was comparable to that of (1,2); nevertheless, the undesirable fluorescence in the single-target scenario [A/TI of (2,0)] was noticeable (green curve, Figure 2C), presumably ascribed to no UN left in TI. The ratio (2,0) between signal (A/TG/TI) and noise (A/TI) was 3.58 (at 47 °C), which approximated to 3.86 in (2,1) (Figure 2D). The results conclude that the formation of the 4WJ requires a certain number of UNs (at least one in each target) for cushion utility. Under the incorporation of the assistant strand, a single target with no UN substantially increases the probability to unzip the MB. In addition, exceeding number of UNs confers 4WJ flexibility, which nevertheless decreases the fluorescence. Thermodynamic Characterization of MB Unfolding. Sensing performance of MB-based sensors is considered to be relevant to its unfolding thermodynamics and kinetics. While unfolding behavior of MB upon hybridization with a perfectly complementary target has been investigated,22,23 no research documented thermodynamics and kinetics of unfolding against dual targets. Equilibrium constants between phase 1 and 2 (K12, Scheme 1B) as well as that between phase 2 and 3 (K23, Scheme 1B) for each UN’s composition were derived (Table 1). The derivation was based on an assumption referring stable Table 1. Equilibrium Constants and Rate Constants of Molecular Beacons upon the Hybridization with Hybrids of the Assistant Strand and Dual Targets Bearing Varied Combinations of Unpaired Nucleotidesa K23 A2-TG-TI (1,1) A4-TG-TI (2,1) A3-TG-TI (2,0) A8-TG-TI (0,0) A9-TG-TI (1,2) A7-H-N (2,0)b A7-H-N (2,0)c A7-H-N (2,0)d

5.41 5.41 5.41 5.41 5.41 4.94 6.96 1.86

± 0.43 × ± 0.43 × ± 0.43 × ± 0.43 × ± 0.43 × × 10−4 × 10−3 × 10−2

k (M−1 s−1)

K12 −4

10 10−4 10−4 10−4 10−4

1.43 1.56 1.37 4.45 1.49 1.71 8.84 1.77

× × × × × × × ×

−6

10 10−6 10−6 10−6 10−6 10−6 10−7 10−7

29400 35900 28700 33200 29500 32000

Results were obtained at 32 °C with MB1 except for the ones indicated otherwise. bDerived at 32 °C with MB2. cDerived at 37 °C with MB2. dDerived at 45 °C with MB2. a

formation of A/TG/TI ensemble (indicated with the red dotted cycle, Figure 3A) to be considered as a single target. This enabled a numerical evaluation in accordance with the derivation illustrated previously.22,23 K23, characteristic of the inherent feature of MB1, was derived to be (5.41 ± 0.43) × 10−4. Among the combinations under investigation, the (2,0) and (0,0) exhibited the smallest (1.37 × 10−6) and greatest K12 (4.45 × 10−6), respectively, in agreement with the relatively high and low fluorescence intensity (shown in Figure 2, parts C and A) reflecting the formation of [MBT]. Other combinations gave K12 ranging to (1.43−1.56) × 10−6. K12 reflects the transition between associated (four-strand complex) and dissociated states in equilibrium ([MBclosed][T]/[MBT]). Low K12 [e.g., (2,0) and (1,2)] indicated the tendency toward association. On the other hand, the greatest K12 in (0,0) can be ascribed to a high tension in the 4WJ as a result of no UNs left, thereby impeding the formation of the four-strand complex. Further insight was gained by evaluating the entropy change ΔS12 (Supporting Information Figure S-2) with summarized results in Table 2. Considering that the triplex ensemble (phase 2, Scheme 1B) in distinct combinations can be stably formed by

Figure 2. Thermal denaturation profiles of MB1 alone (black) and in the presence of the assistant strand and TG (blue), assistant strand and TI (green), as well as assistant strand, TG, and TI (red), under varied combinations of unpaired nucleotides left in the target sequences: (A) (0,0), (B) (1,1), (C) (2,0), and (D) (2,1). The numbers printed in parentheses denote the unpaired nucleotide number left in TG and TI. The concentration of MB1 was 0.2 μM. The concentration for assistant strands, TG, and TI was 1.2 μM. D

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

Figure 3. Polyacrylamide gel electrophoresis showing (A) formation of the triplex ensemble and (B) DNA complex with four-way junction. (A) lane 1, A9/TG/TI; lane 2, A9/TG; lane 3, A9/TI; lane 4, A9; lane 5, TG; lane 6, TI. (B) lane 1, MB2/A7/H/N; lane 2, MB2/H/N; lane 3, MB2/A7/H; lane 4, MB2/A7/N; lane 5, MB2; lane 6, A7; lane 7, H; lane 8, N. Lane M in both panels represents DNA ladder marker.

structure), ΔS12 can be greater than others. As a result, the combination (1,2) had the highest ΔS12 (696 J/mol·K). The sequent ones were (1,1) and (2,0), with ΔS12 of 647 and 643 J/ mol·K, respectively, whereas the (2,1) combination had ΔS12 of 606.51 J/mol·K, in correlation with the smaller fluorescence intensity [14.5% lower than (1,2), shown in Figure 2D]. Unfolding Dynamics of the MB. As shown in Figure 1B, fluorescence rises immensely in the beginning 30 s and then reached a plateau gradually after 50 s. The unfolding rate constant (k) was derived to describe the unfolding kinetics of MB1 quantitatively (Table 1). Notably, in the beginning 10 s after the injection of the target solutions, a slightly greater slope was observed in (2,1) compared with those of (1,1) and (2,0). We attributed the fast unfolding kinetics in (2,1) (k = 35 900 M−1 s−1) and (1,2) (k = 33 200 M−1 s−1) to the increased number of UNs that facilitated the formation of the 4WJ. On

Table 2. Changes of Enthalpies and Entropies for the Unfolding of MB1 upon the Hybridization with the Assistant Strand/Dual-Targets Hybrids in Connection to Varied Combinations of Unpaired Nucleotides A2-TG-TI A4-TG-TI A3-TG-TI A9-TG-TI

(1,1) (2,1) (2,0) (1,2)

measured θm

ΔH12 (kJ/mol)

ΔS12 (J/mol·K)

52 51 52 52

264 234 246 248

647 607 643 696

virtue of the 15-bp duplexes (formed at A-I and A-II), leading to negligible entropy alteration, S2 remained constant for the various combinations. As such, ΔS12 was reversely dependent on the quantity of S1. In the circumstance where the four-strand ensemble (A/TG/TI/MB) was anticipated to prevail in phase 1 and exhibit low entropy (due to its stable and ordered

Figure 4. (A) Thermal denaturation profiles of MB2 solution (black) containing H (dark cyan), N (magenta), H and N (blue), A7 and H (orange), A7 and N (yellow), as well as A7, H, and N (red). Inset: magnified plots for the curves given at temperatures from 30 to 55 °C. (B) Dynamic curves of the MB2 unfolding in the absence (black) and presence of H and N (blue), A7 and H (green), A7 and N (magenta), as well as A7, H, and N (red). The arrow indicates the time point when the assistant strand/targets hybrid was injected. The operating temperature was 45 °C. The concentration of MB2 was 0.2 μM. The concentration of A7, H, and N was 1.2 μM. (C) Temperature-dependent fluorescence recorded at varied target concentrations. (D) Fluorescence (45 °C) recorded as a function of target concentration. E

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

obtain amplicons (see Supporting Information for details). The unpurified amplicons were mixed with MB2 and A7 with resulting fluorescence at 50 °C depicted in histograms (Figure 5). The A/H5/N2 composition demonstrated the greatest

the other hand, the association between MB1 and the target hybrid was potentially retarded in the (1,1) and (2,0) combinations due to less number of UNs. Interestingly, the present system gives greater k (ca. 3−10-fold) than those obtained from varied MBs against a single target,23 inferring a shorter response time, which is beneficial to sensor application. Viability of Strategy against a Genuine Viral Sequence. The dual-target sensing strategy operated with a single MB was further applied to analysis of sequences from the H5N2 virus isolates. For the genuine viral sequence, the selection process (see Supporting Information for details) generated two specific candidate fragments (18- and 16-nt, from H5-coding gene) and four specific fragments (26-, 20-, 17-, and 16-nt, from N2-coding gene) (Supporting Information Table S-2). The longest candidate fragment, in principle, was most desired for the target sequence. However, one should consider the arisen issues including stability of duplexes in the 4WJ, formation of dimer of the target sequence, and unexpected dehybridization at the end base pair of the duplex (Supporting Information Figures S-3−S-5). Eventually, MB2 was designed, which consisted of a 24-nt loop and 10-bp stem. The melting temperature and Gibbs free energy of MB2 were predicted to be 68.2 °C and −5.17 kcal/mol (by mfold Web Server, given at 45 °C with 50 mM Na+ and 5 mM Mg2+). The envisaged 4WJ (Supporting Information Figure S-6) comprises an assistant strand (A7) to hybridize with HA and NA genes with 15 and 14 bp, respectively. The number of UNs on HA was two, yet no UN was designed on the NA side. The gel page (Figure 3B) confirmed the resultant 4WJ construct (red dotted cycle) and unreacted structuration in the H/N, A7/H, or A7/N circumstances (lanes 2, 3, and 4). Analytical Performance of the MB-Based Sensor for Influenza A Virus. For the genuine H5N2 viral sequences, MB2 exhibited paramount fluorescence signal (∼50 °C) only in the concurrent presence of A7, H, and N (red curve, Figure 4A). Albeit the curves given by A7/H and H/N hybrids exhibited fluorescence increase at 42 and 35 °C, respectively, they did not deteriorate the integrity of the AND-gate processing. Unfolding dynamics of MB2 upon treatment of H5N2 sequences were also examined (Figure 4B), along with characteristic thermodynamic and kinetic data summarized in Table 1. To the practical point of view which aims for a welldefined dynamic differentiation of fluorescence and a fast response, varied temperature was studied (Table 1). The dynamic curves at 45 °C are shown in Figure 4B. The fluorescence given by A7/H/N attains an equilibrium state around 1200 s where its intensity approximates 8-fold greater than those given by A7/H and H/N hybrids. For immediate output display, the fluorescence output can be distinguished within 5 min (>4-fold greater than A7/H, A7/N, and H/N), which can be referred as the response time. The fluorescence as a function of target concentration was also investigated (Figure 4C). The relationship is linear over 1.2−240 nM (Figure 4D) in connection to an estimated limit of detection (3σ) at 120 pM (n = 3). Although such limit of detection is insufficient to directly recognize a genetic sequence in clinical samples, the approach is potentially used for subtyping of influenza A virus after PCR amplification or as a part of enzyme-assisted target recycling.24 Detection in Unpurified Amplicon. The system was subsequently applied to analyze PCR product isolated from viral particles [A/duck/Taiwan/DV30−2/2005 (H5N2)]. Asymmetric polymerase chain reaction was performed to

Figure 5. Histogram representing the magnitude of the fluorescence obtained using the AND-gate approach in the four scenarios composed of aPCR products from H5 and N2 genes. Inset: relationship between fluorescence and temperature for the presence of assistant strand A (curve a), A/H5 (curve b), A/N2 (curve c), and A/H5/N2 (curve d). Concentration of MB: 0.2 μM. Concentration of A: 1.2 μM.

fluorescence level (0.26) unambiguously distinguishable from those in other combinations. Consequently, an explicit decision threshold (the horizontal dotted line) can be established, allowing a high-fidelity subtyping of H5N2. The result has shown the capability of the approach to simultaneously recognize two specific sequences in connection to a decisive output. The concept could be readily expanded toward rapid screening of other influenza viruses or any application in which dual-input assessments are demanded, reducing the use of costly molecular beacons, complexity in data interpretation, and requirement for professional personnel. Although MB-based molecular logic gates have been demonstrated previously,25−27 the present work implements a rear application of logic-gate-based sensors to practically significant biomedical targets. In comparison with prior works,28,29 we have accomplished the configuration of a 4WJ, a structure comprising a minimum number of MB and assistant probe, specifically subjected to concurrent identification of multiple genuine sequences from an influenza isolate. The approach provides high selectivity to identify a specified strain (e.g., highly pathogenic or high mortality) in which the screening for both HA and NA are required.30 The method prevails over the conventional protein-based diagnostics which suffer from insufficient specificity against the highly mutated surface antigens HA and NA, and therefore can only qualitatively identify nucleoprotein.31−33 In addition, the present method also potentially outperforms another category of influenza diagnostics which utilizes real-time reverse transcriptase polymerase chain reaction (real-time RTPCR).34−37 Many of the assays ignore the biological significance of NA and often subtype the virus with only HA gene.38,39 When it does for both HA and NA genes, dual outputs are reported (e.g., an H5N1 reagent kit, LightMix from TIB MolBiol, Berlin, Germany) to be interpreted by professional personnel, which is unfavorable for point-of-care diagnostics. The principle of logic processing of dual inputs demonstrated in the present study exactly circumvents the shortcoming. Nevertheless, it should be noted that the design F

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX

Article

Analytical Chemistry

(8) Mackay, I. M.; Arden, K. E.; Nitsche, A. Nucleic Acids Res. 2002, 30, 1292−1305. (9) Marras, S. A.; Tyagi, S.; Kramer, F. R. Clin. Chim. Acta 2006, 363, 48−60. (10) Piatek, A. S.; Tyagi, S.; Pol, A. C.; Telenti, A.; Miller, L. P.; Kramer, F. R.; Alland, D. Nat. Biotechnol. 1998, 16, 359−363. (11) Cornett, E. M.; Campbell, E. A.; Gulenay, G.; Peterson, E.; Bhaskar, N.; Kolpashchikov, D. M. Angew. Chem., Int. Ed. 2012, 51, 9075−9077. (12) Huang, J.; Yang, X. H.; He, X. X.; Wang, K. M.; Liu, J. B.; Shi, H.; Wang, Q.; Guo, Q. P.; He, D. G. TrAC, Trends Anal. Chem. 2014, 53, 11−20. (13) Tan, W.; Wang, K.; Drake, T. J. Curr. Opin. Chem. Biol. 2004, 8, 547−553. (14) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303−308. (15) Klein, D. Trends Mol. Med. 2002, 8, 257−260. (16) Mackay, I. M. Clin. Microbiol. Infect. 2004, 10, 190−212. (17) Lai, Y.-H.; Sun, S.-C.; Chuang, M.-C. Biosensors 2014, 4, 273− 300. (18) Lai, Y. H.; Lee, C. C.; King, C. C.; Chuang, M. C.; Ho, J. A. A. Chem. Sci. 2014, 5, 4082−4090. (19) Hemphill, J.; Deiters, A. J. Am. Chem. Soc. 2013, 135, 10512− 10518. (20) Lai, Y.-H.; Liao, Y.-C.; Mu, J.-J.; Kuo, T.-M.; Hsu, Y.-H.; Chuang, M.-C. Chem. Commun. 2014, 50, 12018−12021. (21) Taubenberger, J. K.; Morens, D. M. Rev. Sci. Tech. OIE 2009, 28, 187−202. (22) Bonnet, G.; Tyagi, S.; Libchaber, A.; Kramer, F. R. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 6171−6176. (23) Tsourkas, A.; Behlke, M. A.; Rose, S. D.; Bao, G. Nucleic Acids Res. 2003, 31, 1319−1330. (24) Gerasimova, Y. V.; Kolpashchikov, D. M. Chem. Soc. Rev. 2014, 43, 6405−6438. (25) Lake, A.; Shang, S.; Kolpashchikov, D. M. Angew. Chem., Int. Ed. 2010, 49, 4459−4462. (26) Genot, A. J.; Bath, J.; Turberfield, A. J. J. Am. Chem. Soc. 2011, 133, 20080−20083. (27) Park, K. S.; Seo, M. W.; Jung, C.; Lee, J. Y.; Park, H. G. Small 2012, 8, 2203−2212. (28) Kolpashchikov, D. M. J. Am. Chem. Soc. 2006, 128, 10625− 10628. (29) Gerasimova, Y. V.; Hayson, A.; Ballantyne, J.; Kolpashchikov, D. M. ChemBioChem 2010, 11, 1762−1768. (30) Uyeki, T. M. Clin. Infect. Dis. 2009, 49, 279−290. (31) Hurt, A. C.; Alexander, R.; Hibbert, J.; Deed, N.; Barr, I. G. J. Clin. Virol. 2007, 39, 132−135. (32) Nidzworski, D.; Pranszke, P.; Grudniewska, M.; Krol, E.; Gromadzka, B. Biosens. Bioelectron. 2014, 59, 239−242. (33) Storch, G. A. Curr. Opin. Pediatr. 2003, 15, 77−84. (34) Ellis, J. S.; Zambon, M. C. Rev. Med. Virol. 2002, 12, 375−389. (35) Stone, B.; Burrows, J.; Schepetiuk, S.; Higgins, G.; Hampson, A.; Shaw, R.; Kok, T. W. J. Virol. Methods 2004, 117, 103−112. (36) Troppan, K. T.; Bozic, M.; Santner, B. I.; Kessler, H. H. J. Clin. Virol. 2010, 49, 82−84. (37) Ginocchio, C. C. Clin. Infect. Dis. 2011, 52, S312−S325. (38) Spackman, E.; Senne, D. A.; Myers, T. J.; Bulaga, L. L.; Garber, L. P.; Perdue, M. L.; Lohman, K.; Daum, L. T.; Suarez, D. L. J. Clin. Microbiol. 2002, 40, 3256−3260. (39) Ng, L. F. P.; Barr, I.; Nguyen, T.; Noor, S. M.; Tan, R. S. P.; Agathe, L. V.; Gupta, S.; Khalil, H.; To, T. L.; Hassan, S. S.; Ren, E. C. BMC Infect. Dis. 2006, 6, 40.

of the sequences of MB and related probes in response to varied targets requires expertise to accomplish. Recognition of expanding number of targets on the basis of a single MB is promising toward versatile analysis, yet design of configuration requires further exploration.

■

CONCLUSIONS In summary, we have implemented a molecular beacon (MB)based sensor which exhibits capability and advantages of recognizing binary targets simultaneously. The approach explores the robustness and versatility of MB as a tool for processing information from binary targets and reporting a decisive conclusion by means of fluorescence resonance energy transfer. Coupling the assistant strand has been shown to be effective in both connecting the binary targets and enabling formation of the DNA 4WJ. The UNs left in the center of the 4WJ exhibited profound but significant impacts on the transition between dissociation (signal “off”) and association (signal “on”) states of the system. As such, effects of the composition of UNs on the equilibrium constant, unfolding rate constant, and change of entropy were evaluated. Taking genuine viral sequences of influenza H5N2 virus as targets, the UNs combination (2,0) in cooperation with the MB containing 10-bp stem and 24-nt loop were designed. Such a system responded to the H5- and N2-specific sequences with a association rate constant of 29 500 M−1 s−1 and an equilibrium constant of 8.84 × 10−7 (37 °C). To the practical point of view, the system exhibits response time less than 5 min, a linear concentration range of 1.2−240 nM, and detection limit at 120 pM. Furthermore, in cooperation with PCR, subtyping of avian influenza viruses was successfully implemented in the sample of unpurified amplicons.

■

ASSOCIATED CONTENT

* Supporting Information S

Additional information as noted in text. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b00810.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +886-4-2359-0121, ext. 32218. Fax: +886-4-23590426. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank Professor Yuan-Hao Hsu for intensive discussion. This work was supported by the Ministry of Science and Technology Taiwan in Grants NSC 102-2113-M-029-003MY2 and NSC 103-2320-B-400-001.

■

REFERENCES

(1) He, L.; Hannon, G. J. Nat. Rev. Genet. 2004, 5, 522−531. (2) Hobert, O. Trends Biochem. Sci. 2004, 29, 462−468. (3) Peter, M. E. Oncogene 2010, 29, 2161−2164. (4) Musser, J. M. Clin. Microbiol. Rev. 1995, 8, 496ff. (5) Nachega, J. B.; Chaisson, R. E. Clin. Infect. Dis. 2003, 36, S24− S30. (6) Sandgren, A.; Strong, M.; Muthukrishnan, P.; Weiner, B. K.; Church, G. M.; Murray, M. B. PLoS Med. 2009, 6, e2. (7) Bell, A. S.; Ranford-Cartwright, L. C. Trends Parasitol. 2002, 18, 337−342. G

DOI: 10.1021/acs.analchem.5b00810 Anal. Chem. XXXX, XXX, XXX−XXX