Recombinase Polymerase Amplification Based Multiplex Lateral Flow Dipstick for Fast Identification of Duck Ingredient in Adulterated Beef

Simple Summary The adulteration and authenticity of meat and meat products has become a global social problem. Beef is often intentionally adulterated with cheap meat. In order to ensure the authenticity of meat, and provide technical support to regulatory authorities, we developed a rapid and visual method to detect duck ingredient in adulterated beef. This method is implemented recombinase polymerase amplification (RPA) and multiplex lateral flow dipstick (MLFD) cascade. The whole RPA-MLFD reaction process can be finished within 35 min, and the results can be determined by naked eyes. RPA-MLFD was applied to simultaneously detect duck ingredient and beef ingredient without using additional instruments. An adulteration ratio as low as 5% of duck ingredient in beef can be easily measured. Moreover, we confirmed that our new method held good potential in the detection of commercially processed meat samples. Therefore, this study reports a useful animal derived meat adulteration detection method, which have potential application in future. Abstract Meat adulteration has become a global social problem. In order to protect consumers from meat adulteration, several methods have been developed to identify meat species. However, the conventional methods are labor-intensive, time-consuming and require instruments. In the present study, a rapid and visual method based on recombinase polymerase amplification (RPA) and multiplex lateral flow dipstick (MLFD) was developed to detect duck ingredient in adulterated beef. Using recombinase and strand displacement polymerase enable RPA to amplify different double-labeled DNA amplicons at room temperature, which can be further detected by MLFD. The whole reaction process can be finished within 35 min, and the results can be determined by naked eyes. As low as 5% of duck ingredient in adulterated beef can be easily measured. Moreover, we confirmed that our new method held good potential in the detection of commercially processed meat samples. In conclusion, this study reported a useful animal derived meat adulteration detection method, which have potential application in future.


Introduction
Recently, the adulteration and authenticity of meat and meat products has become a global social problem. One of the main kind of adulterations in meat and meat products is the partial or total 2.2. RPA Primer and Probe Design DNA sequences were obtained from Ensembl database (http://asia.ensembl.org/index.html). The candidate sequences were tested for homology with other species using online Basic Local Alignment Algorithm Search Tool (BLAST). The genomic sequences for Bovine (Gene number: ARS-UCD1.2:23:10955159:10956296:1) and Anatinae (Gene number: ENSAPLG00000007071) were selected as target sequences. RPA primers and probes were designed by Primer 5.0 software. All primers and probes were synthesized by Shanghai Generay Biotech Co.Ltd. (Shanghai, China). One primer pair and probe for each specie were selected to do further test ( Table 1). The 5 end of reverse primer for beef identification was labeled with biotin, and the 5 end of reverse primers for duck detection was labeled with digoxin. The 5 end of the probes was labeled with fluorescein isothiocyanate (FITC). The C3 blocker were attached to the 3 end of the probe, which could prevent nucleobase from polymerizing during amplification. In addition, the insertion of an abasic furan (dSpacer) which was introduced in the probe sequence to mimic an abasic site.

Preparation of AuNPs and AuNP-FITC Conjugates
Gold nanoparticles (AuNPs) were prepared by sodium citrate reduction method [30]. A total of 200 mL of ultrapure water was continuously heated and rotated to boiling, then 2 mL of chloroauric acid (HAuCl 4 ·3H 2 O, final concentration = 0.01%) and 3.6 mL of 1% trisodium citrate were added until the solution became wine-red. Considering the important role of protein and pH in the binding of AuNPs to protein complexes, the pH and protein content were optimized. After the AuNPs were cooled to room temperature, 15 µL of 0.01 M K 2 CO 3 and 6 uL of monoclonal FITC antibody were added to 1 mL of AuNPs solution to adjust the pH and protein. Subsequently, 1.13 mL of 10% BSA solution and 292 uL of 2% PEG20000 were added and mixed to block the excess reaction sites of AuNPs, 10% sucrose and 0.01% Tween-20 were added to increase the surface activity of AuNPs. Finally, the AuNPs were centrifuged at 4 • C and 2000× g for 30 min, and then the supernatant was removed and the obtained particles were suspended in 1 mL of borax buffer as AuNP-FITC conjugates.

Preparation of Lateral Flow Dipsticks
The LFD was assembled with four parts on a plastic backed sheet in the PVC card, including a sample pad, a conjugate pad, a nitrocellulose membrane, and an absorbent pad ( Figure 1). Firstly, the AuNP-FITC conjugates were evenly coated on the conjugate pad of the dipsticks, then placed in an oven at 60 • C for 15 min to 20 min for drying. Then, monoclonal Biotin antibody and Digoxin antibody were diluted to 1 mg/mL in PBS solution and spotted on a nitrocellulose membrane, generating the beef and duck test line (T line), respectively. The FITC secondary antibody was diluted to 1 mg/mL in PBS solution and spotted on a nitrocellulose membrane to generate the control line (C line). Finally, the nitrocellulose membrane was dried at 37 • C for 30 min. assay. Firstly, the two oligonucleotide primers form a complex with the recombinase proteins. This complex is able invade the target DNA and directs the primer to homologous sequences. A continuous amplification at 39 °C takes place by strand-displacement synthesis catalyzed by a DNA polymerase, while single-strand binding proteins (SSB) stabilize the displaced strand. Then biotinlabeled or digoxin-labeled nucleic acid amplification products could be hybridized with fluorescein isothiocyanate (FITC)-labeled specific probes. The nuclease nfo contained in the reaction solution recognized the abasic site and cleaved the phosphodiester bond to create a free hydroxyl end where DNA polymerase could extend and continue the amplification process. Finally, the dual-labelled amplicon for beef (biotin-RPA product-FITC) and duck (digoxin-RPA product-FITC) were generated. The mixed RPA amplification were dipped into lateral flow dipstick for MLFD detection.

RPA for DNA Amplification
RPA amplification was carried out with the TwistAmp nfo kit (TwistDx, Cambridge, UK). Each reaction was performed in a 23.7 μL reaction mixture containing 15 μL of rehydration buffer, 0.8 μL of forward and reverse primers (10 μM each), 0.3 μL of RPA probe (10 μM), 4.8 μL ddH2O and 2 μL of the DNA template. Then, the reaction mixture was added to the freeze-dried tube in TwistAmp nfo kit. To initiate the reaction, 1.3 μL of 280 mM magnesium acetate (MgAc) was added. The reaction temperature and time were 39 °C and 30 min through the optimization. The obtained RPA products were electrophoresed on a 2% agarose gel stained with GelRed (Biotium, California, CA, USA) in 1 × TAE buffer.

Detection of Processed Meat Samples with RPA-MLFD
For processed meat samples detection, RPA amplifications were carried out with the beef and duck specific primers and probes. Then, the two amplification products were mixed in one tube, and 5 μL of the mixed amplification products were diluted in 55 μL of PBS solution. The dilution was put on the multiplex lateral flow dipstick for 2 to 5 min to observe the results.

Design of RPA-MLFD Assay
In order to fast determine duck ingredient in adulterated beef, the recombinase polymerase amplification combined with multiplex lateral flow dipstick (RPA-MLFD) was developed. The first Firstly, the two oligonucleotide primers form a complex with the recombinase proteins. This complex is able invade the target DNA and directs the primer to homologous sequences. A continuous amplification at 39 • C takes place by strand-displacement synthesis catalyzed by a DNA polymerase, while single-strand binding proteins (SSB) stabilize the displaced strand. Then biotin-labeled or digoxin-labeled nucleic acid amplification products could be hybridized with fluorescein isothiocyanate (FITC)-labeled specific probes. The nuclease nfo contained in the reaction solution recognized the abasic site and cleaved the phosphodiester bond to create a free hydroxyl end where DNA polymerase could extend and continue the amplification process. Finally, the dual-labelled amplicon for beef (biotin-RPA product-FITC) and duck (digoxin-RPA product-FITC) were generated. The mixed RPA amplification were dipped into lateral flow dipstick for MLFD detection.

RPA for DNA Amplification
RPA amplification was carried out with the TwistAmp nfo kit (TwistDx, Cambridge, UK). Each reaction was performed in a 23.7 µL reaction mixture containing 15 µL of rehydration buffer, 0.8 µL of forward and reverse primers (10 µM each), 0.3 µL of RPA probe (10 µM), 4.8 µL ddH 2 O and 2 µL of the DNA template. Then, the reaction mixture was added to the freeze-dried tube in TwistAmp nfo kit. To initiate the reaction, 1.3 µL of 280 mM magnesium acetate (MgAc) was added. The reaction temperature and time were 39 • C and 30 min through the optimization. The obtained RPA products were electrophoresed on a 2% agarose gel stained with GelRed (Biotium, California, CA, USA) in 1 × TAE buffer.

Detection of Processed Meat Samples with RPA-MLFD
For processed meat samples detection, RPA amplifications were carried out with the beef and duck specific primers and probes. Then, the two amplification products were mixed in one tube, and 5 µL of the mixed amplification products were diluted in 55 µL of PBS solution. The dilution was put on the multiplex lateral flow dipstick for 2 to 5 min to observe the results.

Design of RPA-MLFD Assay
In order to fast determine duck ingredient in adulterated beef, the recombinase polymerase amplification combined with multiplex lateral flow dipstick (RPA-MLFD) was developed. The first part of RPA-MLFD is to perform RPA reaction with genomic DNA extracted from beef and duck, Animals 2020, 10, 1765 5 of 9 respectively. After a thermostatic water bath at 39 • C, RPA reaction could amplify target genomic DNA and generate double labelled detectable amplificons ( Figure 1).
The visual interpretation of the RPA products was performed on multiplex lateral flow dipstick, and the principle was illustrated (Figure 1). The RPA products for beef and duck, as well as running buffer were mixed in one reaction tube, then the MLFD was dipped into the mixture. The sample solutions could migrate through the whole stripe, and passed the conjugate pad. Then, the FITC-labeled amplification products were captured by anti-FITC-AuNPs conjugate and generating antigen-antibody-AuNPs complex. The immune complex diffused through the chromatographic membrane until to the test (T) lines, and were captured on T lines. The beef T line contains biotin ligand and captured immune complex derived from beef sample, while the duck T line contains digoxin ligand and captured immune complex derived from duck meat sample. Meanwhile, excess anti-FITC-AuNPs conjugate that were not captured T lines continue to diffuse to the control line (C line) and were captured by the secondary antibody to form a color band (Figure 1).

Specificity and Optimization of RPA-MLFD Assay
The specificity of primers and probes play critical role in the species identification of meat and meat products. To evaluate the specificity of RPA-MLFD, the genomic DNA extracted from beef and duck was tested by RPA-MLFD. Consistent with the expected results, only the positive sample have amplified products for gel electrophoresis detection. The sizes of the RPA products specific to beef and duck were 234 and 226 bp, respectively (Figure 2a,b). For dipstick detection, only the T line appeared in the successful amplification using beef and duck DNA as template, respectively. No amplification was observed for the negative DNA or blank (Figure 2c,d). In order to improve the performance of RPA-MLFD, the most suitable conditions for the RPA-MLFD assay were explored with sets of different temperatures and times for each species. The optimal reaction temperature and reaction time were 39 • C and 30 min ( Figure S1). In addition, when amplification products were carried out with MLFD test, the visualized results for different amplification products were consistent with expectations ( Figure 3). Compared to PCR-based methods, the whole procedure of RPA-MLFD for meat specie identification is more rapid (<35 min), easy to operate as it could be carried out at a low temperature (37-42 • C) without professional equipment, and RPA results can be directly visualized on the dipsticks [28,[31][32][33]. Moreover, the design of RPA amplification is relatively simple, only one pair of primers and one probe are needed to complete amplification, which is more convenient than isothermal LAMP [34,35]. These advantages fully showed that the established method was well adapted for detection of meat adulteration.
Animals 2020, 10, x FOR PEER REVIEW 5 of 9 part of RPA-MLFD is to perform RPA reaction with genomic DNA extracted from beef and duck, respectively. After a thermostatic water bath at 39 °C , RPA reaction could amplify target genomic DNA and generate double labelled detectable amplificons (Figure 1). The visual interpretation of the RPA products was performed on multiplex lateral flow dipstick, and the principle was illustrated (Figure 1). The RPA products for beef and duck, as well as running buffer were mixed in one reaction tube, then the MLFD was dipped into the mixture. The sample solutions could migrate through the whole stripe, and passed the conjugate pad. Then, the FITClabeled amplification products were captured by anti-FITC-AuNPs conjugate and generating antigen-antibody-AuNPs complex. The immune complex diffused through the chromatographic membrane until to the test (T) lines, and were captured on T lines. The beef T line contains biotin ligand and captured immune complex derived from beef sample, while the duck T line contains digoxin ligand and captured immune complex derived from duck meat sample. Meanwhile, excess anti-FITC-AuNPs conjugate that were not captured T lines continue to diffuse to the control line (C line) and were captured by the secondary antibody to form a color band (Figure 1).

Specificity and Optimization of RPA-MLFD Assay
The specificity of primers and probes play critical role in the species identification of meat and meat products. To evaluate the specificity of RPA-MLFD, the genomic DNA extracted from beef and duck was tested by RPA-MLFD. Consistent with the expected results, only the positive sample have amplified products for gel electrophoresis detection. The sizes of the RPA products specific to beef and duck were 234 and 226 bp, respectively (Figure 2a,b). For dipstick detection, only the T line appeared in the successful amplification using beef and duck DNA as template, respectively. No amplification was observed for the negative DNA or blank (Figure 2c,d). In order to improve the performance of RPA-MLFD, the most suitable conditions for the RPA-MLFD assay were explored with sets of different temperatures and times for each species. The optimal reaction temperature and reaction time were 39 °C and 30 min ( Figure S1). In addition, when amplification products were carried out with MLFD test, the visualized results for different amplification products were consistent with expectations ( Figure 3). Compared to PCR-based methods, the whole procedure of RPA-MLFD for meat specie identification is more rapid (<35 min), easy to operate as it could be carried out at a low temperature (37-42 °C) without professional equipment, and RPA results can be directly visualized on the dipsticks [28,[31][32][33]. Moreover, the design of RPA amplification is relatively simple, only one pair of primers and one probe are needed to complete amplification, which is more convenient than isothermal LAMP [34,35]. These advantages fully showed that the established method was well adapted for detection of meat adulteration.

Sensitivity of RPA-MLFD Assay
To assess the sensitivity of RPA-MLFD assay in adulterated beef samples detection, raw meat mixtures containing a various concentration of 5, 25 and 50% duck in beef were performed to test. The results had shown that all samples could be detected by two T lines and found that the 5% duck in beef can be easily detected (Figure 4). We were not pursuing a lower proportion because it was difficult to obtain any commercial profit from the 5% adulteration of low-quality meat in actual manufacturing and sales, which was in good agreement with the data reported in many previous reports [35,36]. In addition, the sensitivity test was also conducted for each species, a dilution series of beef and duck DNA with concentrations ranging from 100 to 0.01 ng were performed. The method was found to be highly sensitive with a detection limit of 0.1ng for beef and duck ( Figure S2a,b), which was equivalent to conventional PCR [37,38]. All the results showed that the RPA-MLFD assay had a great sensitivity without any instrument.

Application of RPA-MLFD in Processed Meat Samples
Finally, in order to confirm the practicability of the established RPA-MLFD method, real commercially processed beef products were detected by RPA-MLFD. The results in Figure 5 reveal that all of the commercial products can be accurately detected with our developed method, demonstrating our method was suitable for the detection of processed meat samples. In addition, changing the functional primer and probe sets in RPA-MLFD assay could easily expand the detection

Sensitivity of RPA-MLFD Assay
To assess the sensitivity of RPA-MLFD assay in adulterated beef samples detection, raw meat mixtures containing a various concentration of 5, 25 and 50% duck in beef were performed to test. The results had shown that all samples could be detected by two T lines and found that the 5% duck in beef can be easily detected (Figure 4). We were not pursuing a lower proportion because it was difficult to obtain any commercial profit from the 5% adulteration of low-quality meat in actual manufacturing and sales, which was in good agreement with the data reported in many previous reports [35,36]. In addition, the sensitivity test was also conducted for each species, a dilution series of beef and duck DNA with concentrations ranging from 100 to 0.01 ng were performed. The method was found to be highly sensitive with a detection limit of 0.1ng for beef and duck ( Figure S2a,b), which was equivalent to conventional PCR [37,38]. All the results showed that the RPA-MLFD assay had a great sensitivity without any instrument.

Sensitivity of RPA-MLFD Assay
To assess the sensitivity of RPA-MLFD assay in adulterated beef samples detection, raw meat mixtures containing a various concentration of 5, 25 and 50% duck in beef were performed to test. The results had shown that all samples could be detected by two T lines and found that the 5% duck in beef can be easily detected (Figure 4). We were not pursuing a lower proportion because it was difficult to obtain any commercial profit from the 5% adulteration of low-quality meat in actual manufacturing and sales, which was in good agreement with the data reported in many previous reports [35,36]. In addition, the sensitivity test was also conducted for each species, a dilution series of beef and duck DNA with concentrations ranging from 100 to 0.01 ng were performed. The method was found to be highly sensitive with a detection limit of 0.1ng for beef and duck ( Figure S2a,b), which was equivalent to conventional PCR [37,38]. All the results showed that the RPA-MLFD assay had a great sensitivity without any instrument. . Sensitivity analysis of RPA-MLFD assay for adulterated meat. Lane 1:100% duck; Lane 2: 50% duck + 50% beef; Lane 3: 25% duck + 75% beef; Lane 4: 5% duck + 95% beef; Lane 5: negative control.

Application of RPA-MLFD in Processed Meat Samples
Finally, in order to confirm the practicability of the established RPA-MLFD method, real commercially processed beef products were detected by RPA-MLFD. The results in Figure 5 reveal that all of the commercial products can be accurately detected with our developed method, demonstrating our method was suitable for the detection of processed meat samples. In addition, changing the functional primer and probe sets in RPA-MLFD assay could easily expand the detection Figure 4. Sensitivity analysis of RPA-MLFD assay for adulterated meat. Lane 1:100% duck; Lane 2: 50% duck + 50% beef; Lane 3: 25% duck + 75% beef; Lane 4: 5% duck + 95% beef; Lane 5: negative control.

Application of RPA-MLFD in Processed Meat Samples
Finally, in order to confirm the practicability of the established RPA-MLFD method, real commercially processed beef products were detected by RPA-MLFD. The results in Figure 5 reveal that all of the commercial products can be accurately detected with our developed method, demonstrating our method was suitable for the detection of processed meat samples. In addition, changing the functional primer and probe sets in RPA-MLFD assay could easily expand the detection strategy to identify other adulterated ingredients accurately and quickly. Furthermore, the RPA-MLFD method requires lesser hands-on time and is easy to perform, the low demand for instruments would make the established method achieve expected on-site detection in future. Therefore, the RPA-MLFD method has a broad application prospect in the authenticity detection of animal derived meat. strategy to identify other adulterated ingredients accurately and quickly. Furthermore, the RPA-MLFD method requires lesser hands-on time and is easy to perform, the low demand for instruments would make the established method achieve expected on-site detection in future. Therefore, the RPA-MLFD method has a broad application prospect in the authenticity detection of animal derived meat.

Conclusions
In this study, a novel assay based on isothermal RPA and multiplex lateral flow dipstick detection was demonstrated. The assay was applied to detect duck ingredient in adulterated beef. In this assay, RPA used recombinase and strand displacement polymerase to amplify a specific target sequence at a constant reaction temperature. Using a special probe structure and nuclease, a doublelabeled DNA amplicon was generated in one reaction, which could be detected on same one dipstick. The detection limit of RPA-MLFD method is 0.1ng for beef and duck, and an adulteration ratio as low as 5% of duck in beef can be easily measured. Comparison with conventional PCR methods, another advantage of RPA-MLFD method is the constant low temperature required for amplification and the fast reaction speed from the start of the reaction to reading of results in less than 35 min. In addition, the analysis of the reaction itself and lateral flow dipstick don't require any or only minimal equipment, which is ideal for field testing. An easy-to-see band on one lateral flow dipstick gave a clear yes/no answer, visible to the naked eye and untrained personnel, which was particularly important in remote areas where there may not be trained workers. Therefore, the RPA-MLFD method could be really applied to the meat authenticity detection on site in future.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1, Figure S1. Reaction temperature and time optimization of beef and duck primers and probes. Figure S2. Sensitivity of RPA-MLFD for beef and duck primers and probes.

Conclusions
In this study, a novel assay based on isothermal RPA and multiplex lateral flow dipstick detection was demonstrated. The assay was applied to detect duck ingredient in adulterated beef. In this assay, RPA used recombinase and strand displacement polymerase to amplify a specific target sequence at a constant reaction temperature. Using a special probe structure and nuclease, a double-labeled DNA amplicon was generated in one reaction, which could be detected on same one dipstick. The detection limit of RPA-MLFD method is 0.1ng for beef and duck, and an adulteration ratio as low as 5% of duck in beef can be easily measured. Comparison with conventional PCR methods, another advantage of RPA-MLFD method is the constant low temperature required for amplification and the fast reaction speed from the start of the reaction to reading of results in less than 35 min. In addition, the analysis of the reaction itself and lateral flow dipstick don't require any or only minimal equipment, which is ideal for field testing. An easy-to-see band on one lateral flow dipstick gave a clear yes/no answer, visible to the naked eye and untrained personnel, which was particularly important in remote areas where there may not be trained workers. Therefore, the RPA-MLFD method could be really applied to the meat authenticity detection on site in future.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2076-2615/10/10/1765/s1, Figure S1. Reaction temperature and time optimization of beef and duck primers and probes. Figure S2. Sensitivity of RPA-MLFD for beef and duck primers and probes.