Simultaneous Detection of Actinobacillus pleuropneumoniae and Haemophilus parasuis in Pig by Duplex Droplet Digital PCR

Actinobacillus pleuropneumoniae ( A. pleuropneumoniae ) and Haemophilus parasuis ( H. parasuis ) are important pathogens of swine, which cause great economic losses in the swine industry and bring great safety hazards to food safety. In this study, we established a duplex Droplet Digital Polymerase Chain Reaction (ddPCR) assay for sensitive detection of A. pleuropneumoniae and H. parasuis simultaneously in pork. The limit of detection of plasmid DNA could reach 2 copies/μL for both A. pleuropneumoniae and H. parasuis ; which were approximately 50- and 20-fold greater sensitivity than Quantitative real-time PCR (qPCR). Both qPCR and ddPCR showed high linearity and positive correlation with standards. The detection results of clinical suspected samples for A. pleuropneumoniae and H. parasuis indicated the positive detection rate of ddPCR (10.7% and 12.5%) was higher than that of qPCR (7.1% and 8.9%). Therefore, the duplex ddPCR assay could be used as an effective quantitative detection to diagnose A. pleuropneumoniae and H. parasuis simultaneously, which may lay a foundation for the development of substitute rapid diagnostic technique.

fever broke out in China [5] happened these years, which caused a serious blow to the swine industry. Then the quality and safety of pork closely related to human health cannot control effectively. It is therefore critical for timely monitoring and prevention of swine epidemics to ensure the safety of pork consumption.
Actinobacillus pleuropneumoniae (A. pleuropheumoniae) is a respiratory pathogen in swine, and it is the pathogenic agent of porcine pleuropneumonia in swine [6]. Haemophilus parasuis (H. parasuis) is the specific pathogenic cause of Glässer's disease in swine, which brings about polyserositis in swine, including pleuritis, peritonitistis and arthritis [7]. Both of them are among the most common bacterial causes of porcine respiratory disease, and often associated with the porcine respiratory disease complex caused by both primary and secondary respiratory pathogens [8]. Although these two pathogens will not affect the food safety directly, the present of A. pleuropheumoniae and H. parasuis in pig can damage the quality and taste of meat, and reduce the edible quality of pork. Furthermore, the porcine respiratory disease is associated with a high incidence rate and fatality rate, which causes substantial losses in the swine industry worldwide and the associated healthcare costs [9][10][11].
In recent years, Enzyme-linked Immunosorbent Assay (ELISA) [12,13], nested Polymerase Chain Reaction (nested PCR) [14,15] and Quantitative real-time PCR (qPCR) [16][17][18] have been developed to detect and serotype H. parasuis and A. pleuropneumoniae with improved sensitivity and efficiency. However, it is difficult to distinguish the mixed infection of these two diseases by clinical symptoms and pathological features in clinical diagnosis. Thus, multiplex PCR detection has been widely used in many fields in recent years, such as monitoring food pathogenic microorganisms [19,20] and detecting veterinary diagnosis [21][22][23]. Moreover, Droplet Digital PCR (ddPCR), the latest version of digital PCR, allows better accuracy of DNA quantification [24]. The ddPCR can detect bacterial pathogens with greater sensitivity, which is crucial for early diagnosis and disease control. At present, with the use of commercial ddPCR system such as the Bio-Rad QX100/200, many studies involving duplex quantitative detection have been carried out [25][26][27]. In this study, we established a duplex ddPCR assay for identifying A. pleuropneumoniae and H. parasuis simultaneously in comparison with qPCR.

Primers and Probes
The primers and probes of qPCR and ddPCR assays were designed by using Oligo Primer Analysis Software And the plasmid pUC57-APP and pUC57-HPS were used as the positive, respectively. The purified recombinant plasmids were quantified (Quibt 3.0; Thermo Fisher Scientific, Waltham, MA, USA), and recalculated to plasmid copies/μL using an online calculator (available at http://scienceprimer.com/copy-numbercalculator-for-realtime-pcr). The recombinant plasmids were serially 10-fold and twofold diluted. Then, the dilutions and plasmids were stored at −20°C and −70°C, respectively.

ddPCR Assay
A duplex ddPCR assay was designed to amplify A. pleuropneumoniae and H. parasuis with primers and probes by using QX200 droplet digital PCR system (Bio-Rad Laboratories, Hercules, CA, USA). In each step of amplification, the optimized 20 μL mixtures of duplex ddPCR reaction containing 10 μL of 2 × ddPCR supermix for probes (no dUTP; Bio-Rad), 800 nM of each primer, 400 nM of each TaqMan-MGB probes, and 1 μL of each DNA template. 20 μL of ddPCR reaction mixtures and 70 μL of Droplet Generation oil for Probes (Bio-Rad) were inserted in an eight-well cartridge to generate the droplets. With the use of a droplet generator (Bio-Rad), each sample was divided into 20,000 water-in-oil nanolitre-sized droplets. After that, the 40 μL generated droplet emulsion was transferred to a new 96-well PCR plate (Eppendorf, Hauppauge, NY, USA), then heat-sealed with a pierceable sealing foil sheet by using the PX1™ PCR plate sealer (Bio-Rad) and amplified in C1000 Touch™ deep-well thermal cycler (Bio-Rad) finally. The optimized cycling conditions of ddPCR were as follows: 95°C for 10 min; then 40 cycles of 94°C for 30 s, 56.3°C for 60 s, and one cycle of 98°C for 10 min, and ending at 12°C.

qPCR Assay
The qPCR assay was performed using ABI QuantStudio 6 Flex realtime PCR system (Thermo Fisher Scientific). The 20 μL qPCR reactions comprised 10 μL of AceQ qPCR probe master mix (Vazyme, Nanjing, China), 1 μL of each DNA template, and the final concentrations of primer and probe were 200 and 100 nM, respectively. The reaction conditions of qPCR were as follows: 95°C for 5 min, followed by 35 cycles at 95°C for 10 s, and 60°C for 34 s.

Specificity and Reproducibility Test of ddPCR
The specificity of the ddPCR assay was validated, and nucleic acid extracts of A. pleuropneumoniae, H. parasuis, PCV2, PRV, S. suis, S. aureus, Salmonella, CSFV, PRRSV and PEDV were tested.
The robustness and reproducibility of the qPCR and ddPCR were determined by serially diluted recombinant plasmids. To evaluate intra-and inter-assay reproducibility, each template was tested in triplicate, and the Standard Deviation (SD) and the Coefficient of Variation (CV) were calculated and analyzed statistically.

Comparison of ddPCR and qPCR Assay
The pUC57-APP and pUC57-HPS which serially 10-fold and twofold diluted were used to compare the sensitivity and accuracy between ddPCR and qPCR. The quantitative agreement between ddPCR and qPCR measurements was access by correlation analysis using standard curves of qPCR and ddPCR.

Clinical Sample Detection
One hundred and forty clinical samples and 50 lung tissue samples were used to evaluate the ability of qPCR and ddPCR. The positive detection rate was compared and evaluated the sensitivity of both methods.

Statistical Analysis
The correlations and regressions analysis of the standard curves from qPCR were analyzed by origin 2017 (OriginLab, MA, USA). The copy number of the initial templates from ddPCR was analyzed by QuantaSoft analysis software (Bio-Rad). Kappa statistics were used to evaluate the agreement of clinical detection results between qPCR and ddPCR.

Primer Annealing Temperature Optimizing
According to the manufacturer's instructions, the optimized annealing temperature of qPCR was 60°C. For ddPCR assay, the annealing temperature gradients test was carried out at the following temperatures: 60°C, 59.4°C, 58.3°C, 56.3°C, 53.9°C, 52°C, 50.7°C, and 50°C. As shown in Figure 1, the optimal annealing temperature of ddPCR was 56.3°C, which could effectively distinguish the signals between fluorescent channels peaked.

Specificity and Reproducibility Test of ddPCR
The CV is used to reflect the robustness and reproducibility of ddPCR. As shown in Tables 1 and 2, CV values of intra-assay variation was 0.54-6.67% for A. pleuropneumoniae and 0.64-6.83% for H. parasuis, and inter-assay variation was 0.28-5.33% for A. pleuropneumoniae and 0.85-4.76% for H. parasuis, which indicated good reproducibility and robustness of ddPCR. The results of specificity test showed the target strains and positive samples were correctly identified and no generating false-positive or falsenegative results were detected. Thus, it confirmed that the primers and probes designed in this study had no non-specific amplification in the genomes on the above pathogens, which indicated that ddPCR had good specificity.

Analysis of Standard Curves and Detection Limits
To assess the accuracy and sensitivity, serial dilutions of recombinant pUC57-APP and pUC57-HPS were tested by both PCR assays (Table 3 and Figure 2). Standard curves and efficacies of qPCR were obtained upon serial 10-fold dilutions. A. pleuropneumoniae and H. parasuis PCR efficacies were 1.064 and 1.143, respectively. The Limit of Detection (LoD) of duplex qPCR was 105 and 45 copies/μL for plasmid DNA of A. pleuropneumoniae and H. parasuis, respectively. In contrast, the LoD for ddPCR detection of plasmid was 2 copies/μL for both A. pleuropneumoniae and H. parasuis. The LoD for ddPCR detection of A. pleuropneumoniae and H. parasuis was 50-fold and 20-fold better than qPCR, respectively. The correlation coefficient (R 2 ) was calculated on the mean value of target copy numbers measured in the sensitivity of ddPCR and qPCR, which is used to determine the linearity over the dynamic range. The results were showed that, in the range of quantification, both qPCR and ddPCR have good linearity (Figure 3).

DISCUSSION
The surveillance of A. pleuropneumoniae and H. parasuis depends on the rapid and accuracy diagnosis. Conventional laboratory methods are less sensitive, laborious and time-consuming. Because of the precision and specificity of ddPCR, especially in the serial dilutions of genomic DNA from both pathogens, ddPCR is an effective method to detect bacterial DNA [28]. In addition, compared with other PCR methods, the difference of ddPCR is that each template molecule is amplified in an independent reaction chamber generated by water droplets emulsified in oil [26]. Each droplet acts as an independent micro-reactor, so it can achieve highly efficient amplification. Therefore, at the single-molecule resolution level, the sensitivity of ddPCR is higher than that of qPCR.  In this study, the serially diluted plasmid was used to measure the sensitivity of qPCR and ddPCR. As a novel molecular method, ddPCR has the advantage of absolute quantification detection of nucleic acids without relying on references or calibration curves. Our duplex ddPCR assay was specific and more sensitive than qPCR in detecting A. pleuropneumoniae and H. parasuis simultaneously. The detection of clinical samples that were identified by ELISA confirms that ddPCR has more accurate sensitivity. However, ddPCR is sometimes not as universal as qPCR due to the multiple partitions generated in the test. Furthermore, ddPCR may provide non-linear results when the initial concentration of the sample is too high. As described in the instruction manual of ddPCR, it is not practical for the detection of high concentration with more than 10 5 copies of pathogens in clinical samples. Therefore, qPCR could be used when the pathogen concentration is very high (>10 5 copies in the added template), while ddPCR is acceptable for the clinical samples with a low concentration (<105 copies/μL for A. pleuropneumoniae and 45 copies/μL for H. parasuis in the added template). Nevertheless, the cost of ddPCR reagents is comparatively high compared with qPCR for conventional diagnostic work, which may restrict its use and makes it hard to spread.
In the scope of our knowledge, this is the first assay for sensitive and simultaneous identification of A. pleuropneumoniae and H. parasuis by qPCR and ddPCR. In related molecular detection studies, limits of detection of A. pleuropneumoniae and H. parasuis were reported. A qPCR assay for detecting H. parasuis showed a sensitivity of LoD was 9.5-0.83 CFU per reaction for the boiling method of DNA extraction and 47.5-0.42 CFU per reaction for the PrepMan Ultra method; and for detecting A. pleuropneumoniae performed an analytical sensitivity of five colony forming units/reaction [18,29]. A loop-mediated isothermal amplification assay for detecting H. parasuis displayed a sensitivity of LoD was 0.2 pg/μL and for detecting A. pleuropneumoniae performed at a sensitivity of 8 CFU per tube [30,31]. A multiplex PCR assay for detecting A. pleuropneumoniae, Pasteurella multocida and H. parasuis showed the minimum detection concentration of 100 DNA copies, 10 DNA copies and 7 DNA copies, respectively [32]. In this study, the LoD (2 copies) by duplex ddPCR for both pathogens was lower than the published results. ddPCR may be useful to detect low concentration of clinical tissue samples of A. pleuropneumoniae and H. parasuis.

CONCLUSION
We established a duplex ddPCR assay for A. pleuropneumoniae and H. parasuis with higher specificity and greater sensitivity to those of qPCR. The duplex ddPCR assay is accurate for detecting both pathogens in serial dilutions. What's more, the applicability of ddPCR is evaluated in clinical samples, which exhibits a higher positive detection rate than qPCR. The established method in this study can be a new approach to simultaneous quantitative detection of A. pleuropneumoniae and H. parasuis, so as to reduce the spread of swine diseases and to certain extent ensure the safety of meat food.

CONFLICTS OF INTEREST
The authors declare they have no conflicts of interest.