Duplex,Detection,of,Vibrio,Cholerae,and,Vibrio,Parahaemolyticus,by,Real-time,Recombinase,Polymerase,Amplification＊

LIAO Lei,SHI Wen,MA Chao,TANG Wen Lian,QIAN Qi Lan,WANG Yu,LI Jiao Jiao,SHEN Jin Yang,JI Jing,MA Jin Ming,and GAO Song

VibriocholeraeandVibrioparahaemolyticusare the two most commonly reported pathogens in seafood[1]. Consuming raw or undercooked seafood contaminated with these twoVibriospecies can cause food poisoning, posing the risk of severe gastrointestinal illness and death[2-3]. Therefore,precise and reliable methods for detectingV.

choleraeandV.parahaemolyticuscontamination in seafood are essential for controlling food safety.

Many molecular detection methods have been developed forV.parahaemolyticusandV.cholerae,including polymerase chain reaction (PCR)-based methods, such as PCR and qPCR, and isothermal amplification methods, such as loop-mediated isothermal amplification and recombinase polymerase amplification (RPA)[1,4,5]. Among them,RPA-based methods are more suitable for on-site applications due to the short detection time and less instrument dependence. An RPA assay coupled with lateral flow dipsticks (RPA-LFD) targeting thetlhgene ofV.parahaemolyticushas been established[6],andgyrBofV.parahaemolyticusandlolBofV.choleraehave been selected as targets for detection by real-time RPA assays[4,5]. The real-time RPA assay reads fluorescence signals along with amplification,which avoids opening the reaction containers as in the RPA-LFD assay; thus, there is less risk of carryover contamination in the assay operating environment.

However, previous studies focused on detecting one pathogen at a time. However, multiplex assays to simultaneously detectV.choleraeandV.parahaemolyticusare needed for complex food samples. Current multiplex detection assays available forVibriospecies have used PCR, qPCR, and RPA-LFD technologies[1,7], but a real-time RPA assay for duplex detection ofV.choleraeandV.parahaemolyticushas not been developed. In this study, we describe a duplex real-time RPA assay to simultaneously detectV.choleraeandV.parahaemolyticus. This assay will provide added convenience for on-site detection of these twoVibriospecies in food supply chains, and the principle can be applied to multiplex detection of other foodborne pathogens.

Twenty bacterial strains were used in this study.The reference strains, includingV.cholerae(ATCC 14100),V.parahaemolyticus(ATCC 17802),Vibrio vulnificus(ATCC 27562),Vibrioalginolyticus(ATCC 17749),Vibrioharveyi(ATCC 43516),Vibriomimicus(MCCC 1A02602),Vibriosplendidus(MCCC 1A04096),Vibrioichthyoenteri(MCCC 1A00057),

Aeromonascaviae(ATCC 15468),Aeromonas hydrophila(ATCC 43414),Aeromonasveronii(ATCC 35622),Bacilluscereus(ATCC 14579),Staphylococcus aureus(ATCC 6538),Salmonellaenteritidis(ATCC 14028), andYersiniaenterocolitica(ATCC 9610) were purchased from the American Type Culture Collection (Manassas, VA, USA) or the Marine Culture Collection of China (Xiamen, China). Two environmental strains ofV.choleraeand three environmental strains ofV.parahaemolyticuswere provided by the Jiangsu Institute of Oceanology and Marine Fisheries (Nantong, China). All strains were confirmed by 16S rRNA sequencing and grown in an Alkaline Peptone Water medium at 37 °C when activated. Genomic DNAs were released by boiling the bacterial cultures at 100 °C for 10 min and used directly as reaction templates.

The primer and probe sequences used in the real-time RPA reactions were derived from previously reported individual real-time RPA assays forV.choleraeandV.parahaemolyticus[4,5]. The probes for the real-time RPA reaction were modified as described by the manufacturer’s instructions in the TwistAmp DNA Amplification Exo kit (TwistDx Inc, Maidenhead, UK). A base at the middle of the probe was substituted with a tetrahydrofuran (THF)group, and twoTbases adjacent to THF were labeled with a fluorophore and a quencher (BHQ1 in this study), respectively. The fluorophore probes forV.

choleraeandV.parahaemolyticuswere FAM(detection wavelength 465-510 nm) and HEX(detection wavelength 533-580 nm), respectively.The sequences of the primers and probes are listed in Table 1. The primers and probes were synthesized by General Biology Co. Ltd. (Anhui, China). The target genes,gyrBandlolB, if present, were amplified by RPA with their corresponding forward and reverse primers in the duplex real-time RPA assay. For each target, the respective probe pairs for the amplified strand and the exonuclease III were cut at THF to release the fluorophore for signaling; therefore, the DNA template from either bacterium would be specifically amplified. SpC3 was used to block unnecessary extension of the strand at the 3" end of the probe (Figure 1).

The real-time RPA reactions followed the manufacturer’s instructions for the TwistAmp DNA Amplification Exo kit. A 15-μL real-time RPA reaction mixture was prepared as follows. To the lyophilized enzyme pellet, 35.4 μL of rehydration buffer, 11.9 μL of nuclease-free water, 2.5 μL of primer (s), and 0.7 μL of the probe (s) were added and mixed uniformly to prepare the premix. Then, 13.25 μL of the premix was removed, and 1 μL of the template and 0.75 μL of MgOAc (280 mmol/L) were added. After brief centrifugation, the reaction mixture was incubated at 39 °C for 4 min, gently mixed, and the fluorescence signal was recorded on the Roche LightCycler 480 II qPCR machine (Basel, Switzerland)with the FAM and HEX channels at 39 °C. The signal was read at 1-min intervals for 26 min.

The optimal concentration ranges of the primers and the probe forV.choleraein the single reaction were examined to determine the primer and probe concentrations in the duplex real-time RPA reaction.Using a ten-fold serially diluted template from 1.0 ×100to 1.0 × 104colony forming units (CFU)/μL, the detection limit forV.choleraein the single real-time RPA reaction was 1.0 × 101CFU/μL (Figure 2A). Using this amount of the template, the concentrations of the primers and probe were decreased proportionally, and the fluorescence signal diminished with the decrease in concentration(Figure 2B). Once the concentration was reduced to 25% of the original, a significant drop in the fluorescence signal was observed. Thus, the primer and probe concentrations were used at 50% of their original concentrations (208 nmol/L final for each primer and 58 nmol/L final for the probe) for detectingV.choleraein the real-time RPA reaction.

Table 1. Primer and probe sequences

Similarly, the optimal concentration ranges of the primers and probe forV.parahaemolyticuswere examined in a single reaction. The detection limit of the single reaction was 1.0 × 102CFU/μL (Figure 2C).As the maximum fluorescence value was moderate at the detection limit, 1.0 × 103CFU/μL was used to test the primer and probe concentration ranges. A significant decrease in the signal at a concentration of 25% of the original was observed (Figure 2D),suggesting that the primer and probe concentrations at 50% of the original were acceptable for detectingV.parahaemolyticus.

As the single reactions of the primer and probe concentrations could be reduced to 50% for both pathogens, the primers and probes at concentrations of 50% of the original (208 nmol/L final for each primer and 58 nmol/L final for each probe) were assembled into the duplex real-time RPA reaction to reach the total concentration of 100%. The sensitivity of the duplex assay for each pathogen was determined. The sensitivity forV.choleraewas 1.0 × 101CFU/μL (Figure 2E). A probit regression analysis (SPSS software; IBM Corp.,Armonk, NY, USA) of the results of eight independent repeats showed that the detection limit was 37 CFU/μL in 95% of the cases (Figure 2F). The sensitivity forV.parahaemolyticuswas 1.0 ×102CFU/μL, and the limit of detection was 99 CFU/μL in 95% of the cases (Figure 2G-H).

The specificity of the duplex assay was confirmed with a series ofVibriospecies and other commonly seen zoonotic and foodborne pathogens, including the 15 reference strains and 5 environmental strains described above. In the specificity test, bacterial DNA was extracted using the TIANamp Bacterial DNA Kit (Tiangen Biotech Co., Ltd., Beijing, China). DNA templates were quantified by Qubit 4 (ThermoFisher Scientific Inc., Wilmington, MA, USA) and normalized to 25 ng/μL. Only the reference and environmental strains ofV.choleraeandV.parahaemolyticusproduced positive signals on the respective fluorescence channels (Figure 2I-J), suggesting good specificity. The specificity of the primers and probes used in this study was established previously[4,5].Here we further confirmed no cross-reaction between the 2 primer-probe sets in the duplex system.

Figure 1. Schematic diagram of the duplex real-time RPA assay.

Figure 2. (A-D) Optimizing the primer and probe concentrations. (E-H) Optimizing the duplex real-time RPA reaction. (I-J) Specificity of the duplex real-time RPA assay.

The duplex real-time RPA assay was validated with 40 clinical samples, including 20 shrimp samples(Litopenaeusvannamei), 15 fish samples (Trichiurus lepturus), and 5 shellfish samples (Pectinidae). The detection results were compared to the qPCR results(Supplementary Table S1 available in www.besjournal.com). The qPCR primer sequences were obtained from previous reports and are listed in Table 1[8,9]. Among the 40 samples, 5 were positive forV.cholerae, 7 were positive forV.parahaemolyticus, and 1 was positive for both bacteria. The results were 100% consistent with the qPCR. The duplex real-time RPA assay was accurate and reliable for simultaneously detectingV.choleraeandV.parahaemolyticus. This assay is ready for onsite detection, as the qPCR instrument used in this study can be replaced by a portable fluorescence detector, e.g., the Genie III Scanner from Suntrap Science & Technology Co. Ltd. (Beijing, China).

Amplification was an important issue to solve when establishing the duplex RPA reaction, as biased amplification of one target would consume more reaction resources and affect the other[10]. Here, we designed the reaction so that the amplicon sizes of the two targets were similar (218 bpvs.168 bp) and carefully optimized the primer and probe concentrations. We first determined that the concentrations of the primers and probes could be decreased to 50% of the initial for bothV.choleraeandV.parahaemolyticus. Based on this result, a molar ratio of 1:1 of theV.choleraeprimer-probe set to theV.parahaemolyticusprimer-probe set was selected to compensate for the potential biased amplification. The duplex real-time RPA assay had the same sensitivity as the single reactions for both pathogens, suggesting that the issue of amplification preference was overcome.

In conclusion, a duplex real-time RPA assay to simultaneously detectV.choleraeandV.parahaemolyticuswas established in this study. The assay exhibited good specificity and sensitivity that reached 37 CFU/μL forV.choleraeand 99 CFU/μL forV.parahaemolyticus. These results were comparable to the sensitivity reported for previous singleplex real-time RPA assays[4,5](V.cholerae: 5 copies/μL standard plasmid;V.parahaemolyticus: 1.0 × 102copies/μL genomic DNA). Validation with clinical samples was accurate and reliable. Importantly, the duplex assay improved detection efficiency. This is the first real-time RPA-based multiplex detection assay available forVibriospecies. This assay provides a convenient choice for on-site detection ofV.choleraeandV.parahaemolyticus, and will guide the development of multiplex detection assays for other foodborne pathogens.

No potential conflicts of interest are disclosed.

We thank Dr. HUI Shen of the Jiangsu Institute of Oceanology and Marine Fisheries (Nantong, China)for providing the bacterial strains and clinical samples.

&These authors contributed equally to this work.

#Correspondence should be addressed to JI Jing,E-mail: jijing@jou.edu.cn; MA Jin Ming, E-mail: majinming@outlook.com; GAO Song, E-mail: gaos@jou.edu.cn Tel/Fax: 86-518-85895781.

Biographical notes of the first authors: LIAO Lei, male,born in 1995, Master Degree, majoring in Pharmacy; SHI Wen, female, born in 1997, Graduate Student, majoring in Pharmacy; MA Chao, male, born in 1996, Master Degree,majoring in Pharmacy.

Received: September 14, 2022;

Accepted: October 8, 2022