The sensitivity test using plasmid sample. In the present study, we first tested the sensitivity of the PCR reagents in the amplification on the traditional PCR instrument using the standard plasmid sample. The plasmid sample was 10-fold serial diluted and was detected on the fluorescence quantitative PCR instrument, Bioer, Hangzhou. As shown in Fig. 2, the result showed that the detection limit of the PCR reaction was 1 copies/μL; and the linear correlation coefficient of its dynamic range and the Ct value was 0.9992. It could hence be concluded that the reagent of the PCR detection met the requirements for on-site rapid screening of positive nucleic acids in terms of the detected sensitivity and dynamic range.
The detection limit testing using plasmid sample. We then chose four principal swine viruses, including PRRSV, PEDV, PRV and PCV2 (positive viral nucleic acid was offered by Nanjing Agricultural University), which are often found in related inspection at the customs, and we synthesized the plasmids of these viruses as positive references. The concentrations of the primers and probes, and the reaction temperatures were optimized by means of the plasmid amplifications on the PCR microfluidic chips. The microfluidic chip had 8 chambers, each chamber accommodating 2 μL of the PCR reaction solution. The volume of the amplification system was reduced that was helpful in the rapid PCR process. The rapid heat transfer kept the activity of the Taq DNA polymerase at the maximum level during the rise and fall of temperature. The microfluidic chip with silicon as its substrate further accelerated the heat transfer during the PCR thermal cycle, because of silicon’s excellent thermal conductivity. It was obvious that the amplification reaction’s fluorescence intensity on the chip presented itself in a typical S-shaped curve, and the negative control chamber, in which sterilized water was used as the template, did not show any amplification, indicating that the specificity of the primers and the probe, and the optimization of the amplification temperature were all appropriate here.
The initial concentration of the four plasmids was 109 copies/μL. All the plasmids were 10-fold serial diluted to the final concentration of 1 copies/μL, and every concentration was tested. As shown in Fig. 3, the sensitivity test of the four positive plasmids were 10 copies/μL for PRRSV, 10 copies/μL for PEDV, 100 copies/μL for PRV and 1 copy/μL for PCV2, respectively, which would meet the requirement for detection in terms of the virus load obtained from the clinical samples of the animals.
The specificity testing using positive nucleic acid. Besides the sensitivity test, we also verified the specificity performance of the PCR detections on the microfluidic chip. Based on the positive nucleic acids of four viruses isolated from the clinical sick pigs, which performed the typical symptoms of reproductive disorders, multi-system failure of weaned piglets, and watery diarrhea. We chose one sample as the target for the specificity test, and the other three samples were used as the control. The PCR amplifications were performed on one chip with one reaction chamber filled with the target sample, the rest of the chambers filled with the other three samples as well as other two porcine-transmitted viral agents including porcine transmissible gastroenteritis virus (TGEV) and procine rotavirus (PRVA). As shown in Fig. 4, the result showed that the four PCR protocols all had a very high specificity, because only the target sample presented a significant amplification curve and the control samples had no amplifications. The fast amplification process of the PCR system contributed not only to the high specificity of the PCR amplification but also the short duration in the non-required temperature zone.
The detection performance of the microfluidic PCR system using positive nucleic acid. Finally, we verified the PCR microfluidic chip detection performance using the positive nucleic acid of swine pathogens. Every two chambers on the chip were used for the reaction of one clinical sample, and one chip accommodated four samples in parallel reactions. At the same time, we also performed the detection for the viruses of four swine diseases on a traditional tube-based PCR instrument for parallel reference. As shown in Fig. 5A, each of the four samples had a significant S curve with a varied Ct value (17.5 for PRRSV, 33.5 for PEDV, 18.5 for PRV and 16.5 for PCV2, respectively). The Ct values of the four swine diseases tested on chip-based and tube-based PCR platform showed no significant differences. The differences in the Ct values were due to the variety in the virus abundance of each of the original positive nucleic acid. The magnitude of the fluorescence intensity increment in the different positive nucleic acid at the end of the reaction was due to the different amplification efficiency of primers and probes for different target gene sequences. However, the Ct value was usually used as a criterion when determining the positive or negative nature of the sample. The sample with a Ct value smaller than 35 could be considered as positive threshold, and the sample with a Ct value greater than 35 needed to be re-examined.