Abstract
Porcine circovirus 3 (PCV-3) is a newly emerging virus that poses a potential threat to the swine industry. We developed a sensitive assay utilizing droplet digital PCR (ddPCR) to detect PCV-3. Specificity of the assay was confirmed by the failure of amplification of DNA of other relevant viruses. The detection limit for ddPCR was 1 copy/μL, 10 times greater sensitivity than TaqMan real-time PCR (rtPCR). Both methods showed a high degree of linearity, although TaqMan rtPCR showed less sensitivity than ddPCR for clinical detection. Our findings indicate that ddPCR might offer faster and improved analytical sensitivity for PCV-3 detection.
Porcine circovirus 3 (PCV-3; family Circoviridae, genus Circovirus) is a newly emerging circovirus type, first identified in the United States in 2016, 7 and then in China, Japan, Russia, Europe, Poland, and South Korea.2,3,5,6,10,11 PCV-3 infections have been reported to be associated with porcine dermatitis and nephropathy syndrome, reproductive failure and respiratory diseases, and systemic inflammation.5,7,14 There is a need to establish a rapid, sensitive, and specific method to detect PCV-3 in pigs.
PCR and enzyme-linked immunosorbent assay (ELISA) are routine laboratory methods used to detect PCV-3. 14 To simplify and improve the efficiency of detection, more molecular techniques have been developed, such as real-time PCR (rtPCR) and loop-mediated isothermal amplification methods. The former is dependent on the relationship of the quantification cycle (Cq) to the standard calibration curve for determination of viral loads, and the accuracy of quantification might be affected by inhibitors. 12 The latter is prone to false-positive results because of nonspecific amplification between primers. 14 Thus, these assays have intrinsic limitations.
Droplet digital PCR (ddPCR) is the third generation of PCR technology enabling absolute quantification of DNA targets without the need to construct a calibration curve as used commonly in rtPCR.4,8 Because of absolute quantification and enhanced detection sensitivity, ddPCR has been used widely in microorganism identification.1,9 We established a specific and sensitive method based on ddPCR for the detection of PCV-3.
To validate the specificity of the rtPCR and ddPCR assays, we tested nucleic acid extracts of field isolates of porcine circovirus 1, 2, and 3, porcine parvovirus (species Ungulate protoparvovirus 1), porcine pseudorabies virus (species Suid alphaherpesvirus 1), classical swine fever virus (species Pestivirus C), and porcine respiratory and reproductive syndrome virus (species Betaarterivirus suid 1) that had been identified by PCR sequencing. Fifty clinical tissue samples including livers, lungs, kidneys, spleens, and lymph nodes were collected from 50 pigs that previously tested positive by ELISA for PCV-3 from 10 pig farms. Another 50 serum samples were collected from 50 healthy pigs confirmed negative by ELISA for PCV-3 from 1 pig farm in China. DNA extracted from serum of specific pathogen–free (SPF) pigs was used as a negative control. Field strains, clinical samples, and SPF pigs were obtained from a commercial company (Yongshun Biological Pharmaceutical, Guangdong, China). Total DNA was extracted (DNA/RNA extraction kit; Tiangen Bio-Tech, Beijing, China), according to the manufacturer’s instructions, and stored at −20°C.
The primers and probes were designed for TaqMan rtPCR and ddPCR assays (Oligo Primer Analysis software; Molecular Biology Insights, Colorado Springs, CO) based on the nucleotide sequences of open reading frame 2 (ORF2; GenBank accession MG745692.1), and synthesized (Sangon Biotech, Shanghai, China). The sequences were as follows: F1 (5′-GGTTCCAACGGAAATGACGTT-3′); R1 (5′-GCCCACAGCTGGCACATAC-3′); and probe (FAM-5′-ATGGTGGAGTATTTCTT-3′-MGB). The plasmid standard was constructed by inserting the complete sequence of the ORF2 gene (645 bp) into a pUC57 vector (Sangon Biotech), according to the manufacturer’s instructions, and was then transformed into Escherichia coli DH5α cells. The purified recombinant plasmid was quantified (4 ng/μL; NanoVue Plus; NanoDrop Products, Wilmington, DE), and then serially diluted. Dilutions and plasmids were stored at −20°C and −70°C, respectively. Standard curves for rtPCR were created using 10-fold serial dilutions of plasmid DNA.
Real-time PCR was carried out (ABI QuantStudio 6 Flex real-time PCR system; Thermo Fisher Scientific, Waltham, MA). For each amplification step, the optimal reaction mixtures consisted of 10 μL of AceQ qPCR probe master mix (Vazyme, Nanjing, China), the primers and probe at final concentrations of 400 nM and 200 nM, respectively, 1 μL of template, and double-distilled water to complete a reaction volume of 20 μL. Amplification programs were as follows: 95°C for 5 min, followed by 40 cycles of 95°C for 10 s, and 56.3°C for 34 s.
For the ddPCR system, the same rtPCR primers and probe were used (QX200 droplet digital PCR system; Bio-Rad Laboratories, Hercules, CA). For each amplification step, the optimized ddPCR reaction mixtures included 10 μL of 2× ddPCR supermix for probes (no dUTP; Bio-Rad), 1 μL of template, and the primers and probe at final concentrations of 400 nM and 200 nM, respectively, in a final volume of 20 μL. All reaction mixtures were loaded into a disposable plastic DG8 cartridge (Bio-Rad) with 70 μL of droplet generation oil (Bio-Rad); the mixtures were then placed onto a droplet generator (Eppendorf, Hamburg, Germany) that partitioned each sample into 20,000 nanoliter–sized droplets. Droplets were transferred into a 96-well plate (Eppendorf) for PCR. The reaction programs of ddPCR were as follows: 95°C for 10 min, followed by 40 cycles of 94°C for 30 s and 56.3°C for 60 s, 1 cycle of 98°C for 10 min, and ending at 12°C. Finally, plates containing the droplets were placed into a droplet reader (Bio-Rad); QuantaSoft software (Bio-Rad) provided with the ddPCR system was used for data acquisition to calculate the concentration of target DNA in copies/μL from the fraction of positive reactions.
Correlation and regression analyses of standard curves from rtPCR and ddPCR were performed (Prism software; GraphPad Software, La Jolla, CA). For ddPCR, Poisson statistics were used to measure the initial template concentration (QuantaSoft software). Kappa statistics were used to determine the agreement between ddPCR and TaqMan rtPCR. To avoid generating false-positive results, we monitored for potential contamination throughout the study.
Selection of an optimal annealing temperature and primer concentration was critical for the specificity of a reaction. To select an optimal annealing temperature, plasmid (2.71 × 104 copies/μL) was annealed at the following temperatures: 60, 59.4, 58.4, 56.3, 53.9, 52, 50.7, and 50.0°C. According to the FAM (6-carboxyfluorescein) signals that we displayed as rain plots (Fig. 1), 56.3°C was selected as the optimal annealing temperature, which resulted in a distinct fluorescence amplitude difference between the positive and negative controls and product-enriched droplets. The primer concentration was optimized using pUC57-PCV-3 (2.71 × 104 copies/μL) per reaction mixture. The optimal primer and probe concentrations were 400 nM and 200 nM per reaction mixture, respectively, which had the highest fluorescence amplitude of positive with the same negative droplet numbers (Fig. 2). The results were analyzed using QuantaSoft software.
Fluorescence amplitude of different annealing temperatures for porcine circovirus 3 digital droplet PCR. The assay was conducted across an annealing temperature gradient: 60, 59.4, 58.4, 56.3, 53.9, 52.0, 50.7, and 50.0°C. NTC = no template control.
The fluorescence amplitude of different primer concentrations for porcine circovirus 3 droplet digital PCR. The assay was conducted across a primer and probe concentration ratio gradient: 900:200, 800:200, 400:200, and 300:200. NTC = no template control.
All trials identified the target strains and positive samples correctly without generating false-positive or false-negative results, thereby confirming assay specificity. To compare analytical sensitivity, linearity, and quantitative agreement between ddPCR and rtPCR amplification systems, serial dilutions of recombinant pUC57-PCV-3 were prepared in triplicate, and standard curves were constructed using ddPCR and rtPCR (Table 1, Fig. 3). The limit of detection (LOD) was 1 copy/μL by ddPCR. In contrast, the LOD was 20 copies/μL by rtPCR. Both ddPCR and rtPCR exhibited good linearity, with R2 values of 0.999 and 0.992, respectively (Fig. 3). PCR efficiency was calculated to be 95.4% for ddPCR and 101.3% for rtPCR. Plasmid standards with different copy numbers were used to evaluate the robustness and reproducibility of the ddPCR assay. Each sample was tested in triplicate to evaluate intra- and inter-assay reproducibility. The intra- and inter-assay coefficients of variation (%) for concentration (copies/μL) were 0.09–4.80% and 0.20–4.64%, respectively (Table 2).
Comparison of real-time PCR and droplet digital PCR (ddPCR) using serially diluted porcine circovirus 3 (PCV-3) plasmids.
Cq = quantification cycle; ND = not detected; NTC = no template control.
Concentration based on 10-fold and 2-fold serial dilutions of plasmid.
Concentration based on ddPCR detection.
Quantification of serially diluted porcine circovirus 3 (PCV-3) plasmids by droplet digital PCR (ddPCR) and real-time PCR (rtPCR).
Robustness and reproducibility of the porcine circovirus 3 droplet digital PCR assay.
CV = coefficient of variation; SD = standard deviation.
The 50 PCV-3 antibody-positive tissue samples and 50 negative serum samples were used to compare the detection sensitivities of the ddPCR and rtPCR (Table 3). For negative samples, both rtPCR and ddPCR assays were negative for PCV-3. For the 50 PCV-3 antibody-positive tissue samples, 39 (78%) and 41 (82%) were positive for PCV-3 by rtPCR and ddPCR, respectively. All of the samples correctly identified by rtPCR were also detected by ddPCR. The ddPCR assay was more accurate than the rtPCR assay. The kappa statistic was 0.875, indicating almost perfect agreement between ddPCR and TaqMan rtPCR. Compared with rtPCR, ddPCR has the following advantages: it is capable of absolute quantification without standard curve with excellent reproducibility, 12 it is more sensitive for low copy number quantification, and less susceptible to PCR inhibitors, 13 and thus can detect target DNA in complex environments. These features make ddPCR a promising detection technique and a practical method for clinical diagnosis.
Comparison of sensitivity by droplet digital PCR (ddPCR) and TaqMan real-time PCR for porcine circovirus 3 clinical samples.
Real-time PCR and ddPCR exhibited excellent linearity and efficiency, and were strongly correlated in our study, but the LOD of ddPCR was ~10 times lower than rtPCR. Two clinical samples tested negative by rtPCR but were positive by ddPCR. Our data suggest that the established ddPCR is a more effective method for the precise quantification of PCV-3 compared to the rtPCR system, especially in detecting lower concentrations of PCV-3. This more sensitive method could provide a valuable tool for detecting latent infections in pigs.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This work was supported by the National Key Research and Development Plan (2016YFD0500600), the Guangdong Provincial Science and Technology Plan Project (2017B020207004), the Fundamental Research Funds for the Central Universities (21618309), and the Guang Dong Agricultural Science and Technology Innovation and Apromotion (2018LM2172).