Bacterial and viral strains that had been isolated from clinical specimens were used to validate the TAC. Briefly, the following 20 viruses and 8 bacteria were collected from other laboratories: influenza type A virus, influenza type B virus, enterovirus, parainfluenza virus (subtypes 1 to 3), respiratory syncytial virus (subtypes A and B), human metapneumovirus (subtypes A and B), adenovirus, rhinovirus, human bocavirus, human coronavirus (subtypes HCOV-229E, HCOV-NL63, HCoV-HKU, and HCOV-OC43), measles virus, mumps virus, rubella virus, Mycoplasma pneumoniae, Chlamydia pneumoniae, Coxiella burnetii, Legionella pneumophila, Haemophilus influenza, Mycobacterium tuberculosis, Streptococcus pneumonia, and Bordetella pertussis.
Clinical specimens were obtained from the Affiliated Hospital of the Academy of Military Medical Sciences and the Center for Disease Control and Prevention of Shunyi District in Beijing, China. The protocol for collecting and handling of clinical specimens had been approved by the institutional review boards of both institutes. In total, 754 clinical specimens were collected between July 2013 and May 2015 from patients presenting respiratory syndromes including fever, cough, and runny nose. Nasopharyngeal/oropharyngeal (NP/OP) swabs were kept in 1 mL of universal transport medium and stored at -80°C. All of the 754 specimens were both tested by bacterial culture and individual real-time PCR. 370 were positive, which most of them were identified by real-time PCR, and 384 were negative.
Total nucleic acid extraction
Total nucleic acid was extracted from the clinical specimens (NP/OP swabs) and the bacterial and viral isolates using QIAamp cador Pathogen Mini Kit following the manufacturer’s instructions. Briefly, every sample was pretreatment with buffer ATL and pathogen Lysis L with glass beads according to the manufacturer’s instruction. 200 μL of the medium containing NP/OP swabs was mixed with lysis buffer containing 20 μL proteinase K and 100 μL buffer VXL with 1 μg carrier RNA and incubated at 25°C for 15 min. Total nucleic acid from each sample was eluted in 80 μL buffer AVE. The extracted total nucleic acid was aliquoted and stored in -80°C for future use.
Standard plasmids were used to estimate the amplification efficiency and the limit of detection of the TAC assay. Plasmids containing 30 target genes, including 20 viral and 8 bacterial genes, and 2 internal positive controls were constructed and used as standard plasmids. The 28 target genes were amplified by PCR and the primers are list in table 1. The amplified segments were inserted to the plasmid using the P-EASY-Blunt Zero Cloning Kit (Tran, Beijing, China). Recombinant plasmids were verified by sequencing. The RNA templates were first reverse-transcribed into cDNA, and then amplified and inserted into the plasmid of the cloning kit. The constructed plasmids were purified using the TIANprep Mini Plasmid Kit (Tran) according to the manufacturer’s instructions. The DNA concentration and quality of the purified recombinant plasmids was determined using a NanoDrop 2000 (Thermo Scientific, Waltham, MA). The number of copies per μL of each standard plasmid was calculated according to the equation as described previously (25).
Conventional detection methods
All clinical samples were tested by bacterial culture and individual real-time PCR. The sample was considered positive at least one of the methods give the positive result. Moreover, in order to avoid contamination, all experiments involving conventional detection methods were performed at the Affiliated Hospital of the Academy of Military Medical Sciences and the Center for Disease Control and Prevention of Shunyi District in Beijing, China.
Primer and probe design for TAC and TAC reaction conditions
Primers and probes for 30 target genes (Fig. 1), designed using Primer Express 3.0 or adopted from published reports, were spotted onto the TAC. The primers and probes were designed to detect 28 respiratory pathogens, including a few subtypes, e.g., respiratory syncytial viruses (subtypes A and B), human metapneumovirus (subtypes A and B), parainfluenza virus (subtypes 1 to 3), and human coronavirus (subtypes HCOV-229E, HCOV-NL63, HCoV-HKU, and HCOV-OC43). Highly conserved regions were identified by multiple sequence alignment. To improve the specificity, we labeled probes with minor groove binders instead of black hole quenchers, which are commonly used in commercial TACs. The fluorophore of all probes are FAM. The target genes and the resources of primers and probes are displayed in Table 1.
The final primer and probe concentrations on the plate were 900 nM and 250 nM, respectively. The quantitative One Step qRT-PCR Kit (Tiangen, Beijing, China) was used and each 100-µL reaction mixture contained 50 µL 2× Quant One Step Probe qRT-PCR master Mix, 4 µL HotMaster Taq polymerase, 2 µL QuantRTase, 20 µL DNA/RNA, and 24 µL RNase-free water. The thermal cycling conditions were: 50°C for 30 min, 92°C for 3 min, and 40 cycles of denaturation at 92°C for 10 s, annealing at 62°C for 20 s, and elongation at 68°C for 20 s. The PCR reactions were completed in a ViiA7 real-time PCR instrument (Life Technologies).
Assessment of the amplification efficiency and detection limit of the primers and probes
Linearity and amplification efficiency were assessed as described previously (1, 21). A 10-fold serial dilution of each target gene-containing standard plasmid (107 to 103 copies/µL) was prepared. Standard plasmids at two concentrations (high and low) were tested in triplicate to assess the intra- and interassay variability. The coefficient of variance (CV) was calculated based on cycle threshold (Ct) values. The lowest detection limit (LOD) was defined as the lowest detectable concentration of standard plasmids. When ≥ 5 out of 7 replicates of a standard plasmid at 102 copies/µL to 100 copies/µL were detected, the test was considered positive.
Evaluation of the TAC assay using clinical specimens
Out of the 754 clinical specimens, 370 tested positive and 384 negative in the gold-standard tests. All the negative samples were first processed on the ViiA7 to prevent contamination. RNase-free water was added to each card as a negative control template. A Ct cut-off value of 36 cycles was used to differentiate between positive and negative samples for all clinical specimens and the negative control on the TAC. The sensitivity and specificity of the TAC were evaluated by comparing the results from TAC assay with those from the gold-standard methods. The gold-standard methods were assumed to have 100% sensitivity and specificity. PCR products of 18 randomly chosen positive clinical specimens were sequenced to verify the accuracy of the TAC.
Repeatability (three replicates within one card) and reproducibility (three replicates between cards) of the TAC are shown as CV values. Cohen’s κ was calculated to estimate the degree of consistency between TAC and the gold-standard methods by SPSS 17.0 software. Cohen’s κ was interpreted as follows: < 0, poor; 0–0.20, slight; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; and 0.81–1.00, almost perfect agreement (24). P < 0.05 was regarded statistically significant.
The study was approved by the Fifth Medical Centre, Chinese PLA General Hospital and the Center for Disease Control and Prevention of Shunyi District in Beijing, China. Written informed consent was obtained from each patient or a guardian.