An enzyme free fluorescence resonance transfer strategy based on hybrid chain reaction and triplex DNA for Vibrio parahaemolyticus

In this work, an enzyme-free fluorescence resonance energy transfer (FRET) strategy was established for rapid and specific detection of the DNA sequence from Vibrio parahaemolyticus (VP) using hybridization chain reaction (HCR) amplification and triplex DNA. The triplex forming oligonucleotide (TFO) was labelled with carboxyfluorescein (FAM) as fluorescence donor, and hairpin sequence H1 was labelled by tetramethylrhodamine (TAMRA) as fluorescence receptor. In the present target VP DNA, the hairpin structure of molecular beacon (MB) was opened, the free end was released and hybridized with H1-TAMRA, and the HCR reaction was triggered by the alternate supplementation of H1-TAMRA and H2 to produce the notch double helix analogue. After the addition of TFO-FAM, a triplex structure was formed between HCR products (H1-TAMRA/H2) and TFO-FAM. A close contact between the donor and the receptor resulted in FRET. Under the optimal conditions, the fluorescence quenching value was inversely proportional to the concentration of target VP DNA in the range of 0.1–50 nmol L−1, and the detection limit was 35 pmol L−1.


Scientific Reports
| (2020) 10:20710 | https://doi.org/10.1038/s41598-020-77913-2 www.nature.com/scientificreports/ the donor emission wavelength. FRET efficiency is closely correlated with the spatial distance between donor and recipient molecules 27 . Davies and Rich discovered that when a triplex forming oligonucleotide (TFO) binds to a specific region of dsDNA, the triplex nucleic acid (triplex DNA) can be formed. Homopurine-homopyrimidine dsDNA can be identified by TFO via Hoogsteen (or reverse-hoogsteen) hydrogen bonds. As a powerful tool, triplex DNA interaction plays an important role in regulation of gene expression, construction of molecular switches for biosensors, sequence-specific recognition of the natural state of dsDNA and regulation of DNA nanostructures 28 .
Herein, an enzyme-free fluorescent sensing platform for VP DNA based on HCR amplification, FRET and triplex DNA has been developed. In the presence of VP DNA, which can hybridize with molecular beacon (MB), and the hairpin structure of MB is released. The free terminal of MB can trigger the HCR reaction, and H1-TAMRA and H2 are alternately supplemented to form long nicked dsDNA. The HCR products can be recognized by TFO-FAM and form triplex DNA. So the TAMRA and FAM are brought closer together by the formation of triplex DNA, resulting in a FRET process and a dramatic decrease in the fluorescence intensity at 520 nm.

Materials and methods
Materials and chemicals. All oligonucleotides were synthesized and purchased from Sangon Biotech Co., Ltd. (Shanghai, China), the sequences were listed in Table 1. Oligonucleotides were dissolved in PBS buffer solution (pH 6.0, 10 mmol L −1 , 0.1 mol L −1 NaCl) and stored at 4 °C. Spermine was purchased from Aladdin reagent Co., Ltd. (Shanghai, China). Human serum was purchased from guangzhou hongquan biotech. Co., Ltd. (Guangzhou, China).

Apparatus.
Fluorescence intensity performed with an F-7000 fluorescence spectrophotometer (Hitachi, Japan). The excitation wavelength used was 480 nm, and emission spectra of 490-650 nm were recorded. Gel images were obtained using the ChampChemi Gel Image Analysis System (Beijing, China). Circular dichroism (CD) spectroscopy was analyzed by Chirascan™ V100 spectropolarimeter of Applied Photophysics Ltd. (Britain).

Detection of target VP DNA.
To prepare the hairpin probes (MB, H1-TAMRA and H2), the oligonucleotides were dissolved in PBS (10 mmol L −1 PBS, 0.1 mol L −1 NaCl, pH 6.0), then heated in a water bath at 95 °C for 5 min and cooled at room temperature for at least 4 h. VP DNA was mixed with MB (50 nmol L −1 ) hairpin probe and heated at 85 °C for 5 min and cooled at room temperature for 25 min. H1-TAMRA (100 nmol L −1 ) and H2 (100 nmol L −1 ) were added to the above mixture and incubated at 37 °C for 2 h. TFO-FAM (250 nmol L −1 ) was added to the above reaction mixture and incubated at room temperature for 1 h. Finally, the reaction mixture was detected with a fluorescence spectrophotometer at 480 nm.
Gel electrophoresis analysis. Agarose gel electrophoresis was used to evaluate the HCR process. Agarose gel concentration was 3%, electrophoresis buffer was 1 × Tris-borate-EDTA buffer (TBE, pH 8.0), and constant pressure electrophoresis was performed at 100 V for about 60 min at room temperature. The nucleic acid was stained by 4S Red Plus Nucleic Acid Stain. CD spectroscopy measurement. CD spectroscopy was used to evaluate the structure of DNA and to demonstrate the formation of triplex DNA. The conditions of the spectropolarimeter used for detection were as follows: the scanning range was 200-350 nm, the scanning speed was 10 nm/min, the path was 0.1 cm, and the response time was 1.0 s, and the bandwidth was 1.71 nm.
Preparation of serum samples. Human serum was stored at − 20 °C and thawed at 4 °C before use.
10% human serum was prepared by mixing human serum with PBS buffer (v/v = 1:9) and then filtered with a microporous membrane filter (aperture 0.22 μm). Afterwards, 21.80 nmol L −1 , 27.52 nmol L −1 and 38.88 nmol L −1 VP DNA serum samples were obtained by adding a certain amount of VP DNA to 10% serums. When target VP DNA is absent, HCR cannot be initiated. Resultantly, triplex DNA cannot form, and FAM is too far from TAMRA to FRET. In this case, the fluorescence intensity at 520 nm is strong under 480 nm excitation. With the addition of target VP DNA, the hairpin structure of MB is opened and the free end is released and hybridized with H1-TAMRA, the HCR reaction is triggered by the alternate supplementation of H1-TAMRA and H2 to produce dsDNA. After the addition of TFO-FAM, a triplex structure is formed between HCR products (H1-TAMRA/H2) and TFO-FAM. In this state, FRET pairs (FAM and TAMRA) approach each other, producing higher FRET efficiency. Therefore, quantitative detection of VP DNA can be achieved by detecting the decrease in fluorescence intensity of FAM at 520 nm.
Feasibility of the strategy. To demonstrate the feasibility of the proposed strategy, the fluorescence signal of the system under different conditions were investigated. As shown in Fig. 2, there is no absorption peak at 520 nm due to the lack of TFO-FAM (curve c). When VP DNA was absent, HCR could not be triggered, resulting in the failure of the subsequent process, so the fluorescence intensity at 520 nm was very strong (curve a). In the absence of H1-TAMRA and H2, the double-strand and triplex structures could not form, resulting in FAM and TAMRA far away and FRET could not occur, so the fluorescence was also strong (curve b). However, target VP DNA addition to the mixture of MB, hairpin probe (H1-TAMRA + H2) and TFO-FAM reduced significantly the fluorescence intensity (curve d), indicating that the triplex DNA was formed and FRET occurred.  www.nature.com/scientificreports/ In addition, since target VP DNA activating HCR is prerequisite for building an amplification fluorescence sensing platform, the HCR process before and after the addition of the VP DNA was investigated with gel electrophoresis. As can be seen from Fig. 3, only one low-molecular-weight band in lane 1 (H1 + H2) and two lowmolecular-weight bands in lane 2 (MB + H1 + H2) were observed, indicating that the hairpin probes in the mixed state were stable when VP DNA was missing, however, extended dsDNA appeared after addition of target VP DNA (lanes [3][4][5]. The data showed that the HCR process occurred. Furthermore, it was found that the average molecular weight of HCR products was inversely proportional to the concentration of VP DNA, indicating that the hairpin probe consumption was positively related to VP DNA concentration. In order to demonstrate that HCR products form triplex DNA via binding with TFO, the DNA structures of different mixtures were determined by CD spectroscopy. As shown in Fig. 4, a positive peak at 278 nm and a negative peak at about 248 nm were related to dsDNA appearance in the CD spectrum. The values in curve c increased at 278 nm and about 248 nm as compared with the curve a and b, which was due to the long dsDNA   Figure 6 showed that the fluorescence intensity at 520 nm gradually decreased with the increase of VP concentration, but gradually increased at 590 nm (a). As shown in, with the increase of target VP DNA concentration, fluorescence quenching value at 520 nm gradually increased (Fig. 6b). Meanwhile, the fluorescence quenching value was proportional to the concentration of target VP DNA during the range from 0.1 to 50 nmol L −1 with a linear regression equation of ∆F = 1.83C + 13.61 (R 2 = 0.980, C represents the concentration of VP DNA (nmol L −1 ), ∆F denotes the change of fluorescence intensity) and a detection limit of 35 pmol L −1 (N = 3, RSD = 5.7%). HCR amplification and triplex DNA assembly were the reasons for the high sensitivity of this method. The selectivity of the strategy was studied with different initiators, including single base mismatched (MT1), two base mismatched (MT2) and tree base mismatched (MT3) VP DNA sequences and non-complementary DNA (NCT) (Fig. 7). The addition of NCT did not cause significant changes of fluorescence intensity, while the addition of VP DNA significantly increased the fluorescence quenching value. In addition, the fluorescence quenching value decreased with the increase of base mismatch. These results suggested that the addition of non-complementary DNA failed to open the hairpin structure, and that the ability to open the hairpin structure and triggering HCR decreased with the increase of mismatch bases. Compared with VP DNA, MT1, MT2, MT3 www.nature.com/scientificreports/ and NCT all showed significant differences (p < 0.05 or p < 0.01). So there seems to be that specific sequence of initiator is crucial for HCR initiation, and the HCR based approach has potential in single nucleotide polymorphism (SNP) analysis.
Application. In order to verify the feasibility of this method for the direct quantitative determination of VP DNA in complex biological fluids, VP DNA in human serum was analyzed. Different concentrations of target VP DNA were added to the 10% serum (v/v) for detection according to this proposed method. The results are shown in Table 2. Data showed that this proposed method was feasible to detect target VP DNA in biological samples.

Conclusion
In conclusion, an enzyme-free fluorescence sensitive detection platform based on HCR amplification, FRET and triplex DNA was constructed for sensitive and selective detection of VP DNA. In the presence of VP DNA, the MB hairpin structure is opened to trigger the HCR process to produce dsDNA. When TFO-FAM was added, TFO-FAM hybridized with the HCR products to form a triplex DNA, and the donor and receptor on TFO-FAM and H1-TAMRA are close to each other, resulting in a FRET process. The value of fluorescence quenching was used to quantitatively detect target VP DNA with a limit of detection of 35 pmol L −1 . This method has a broad application prospect in the field of DNA diagnosis and clinical analysis.