Development of a diagnostic assay by three-tube multiplex real-time PCR for simultaneous detection of nine microorganisms causing acute respiratory infections – Scientific Reports

Ethical approval

This study was institutionally approved by Guangdong Pharmaceutical University, the Medical Ethics Committee of the First Affiliated Hospital of the University, and the Medicine and Biological Engineering Technology Research Center of the Ministry of Health, Guangzhou, China, in correspondence with the guidelines of the collection of patient specimens and the use of human biological samples for scientific analysis, as well as data acquisition and presentation. Informed written consent was provided by the participants (or parents, in the case of a minor) involved in the study. In accordance with the ethical requirements, the analysis of all clinical samples was performed in an anonymized manner. All experimental protocols were carried out in accordance with the relevant guidelines and regulations.

Viral and bacterial strains

A total of 25 viral and bacterial strains, including diverse subtypes of the target respiratory viral and bacterial pathogens, used as positive controls for the optimization of multiplex PCR and validation of the sensitivity of this real-time PCR assay, were obtained-listed in Table 5. All pathogenic microorganisms were stored at − 80 °C until DNA/RNA were extracted for analysis.

Table 5 Pathogenic microorganisms, including culture strains and laboratory isolates, used as the target pathogens in the evaluation of three-tube multiplex real-time RT-PCR.

Clinical specimens

Nasopharyngeal/throat swabs were collected in universal transport medium (Supplied by Diagnostic Hybrids, USA) from 179 patients primarily diagnosed with acute respiratory infections. Samples were collected from the respiratory ward of the First Affiliated Hospital of Guangdong Pharmaceutical University within the first 24 h of each patient admission between November 2017 and July 2019. All patients exhibited clinical manifestations of acute respiratory infections with at least two of the following symptoms: cough, pain, soar throat, expectoration and tachypnoea accompanied by fever (temperature > 38 °C). The specimens were collected by specialized professionals and sent to the central laboratory of the hospital for microbiological and immunological analysis. 500 µl aliquots of nasopharyngeal/throat swab samples were sent to our laboratory for the use in the validation of the multiplex real-time RT-PCR.

DNA and RNA extraction

The DNA from 9 subtypes of adenovirus and 7 bacteria and the RNA from a total of 9 subtypes of the target viruses, were extracted using a Nucleic Acid Isolation Kit and a Smart32 Nucleic Acid Extraction instrument, (both supplied by Da’an Gene, Guangzhou, China) in accordance with the manufacturer’s instructions.

The clinical nasopharyngeal/throat swab samples were 500 µl; 200 µl of the mixture was aliquoted for nucleic acid extraction using the same total nucleic acid kit and nucleic acid extraction instrument from Da’an Gene, in line with the corresponding section of the manufacturer’s handbook. The DNA was eluted in a final volume of 70 µl and stored at − 80 °C.

Design of primer–probe pairs for multiplex real-time RT-PCR

Ten sets of primer–probe designed for multiplex real-time RT-PCR are specified in Table 6. Specific primers and probes for the selected targets, nine pathogens plus one internal reference, were designed based on DNA sequences from the International Nucleotide Sequence Database Collaboration (INSDC) at the National Center for Biotechnology Information (NCBI), by utilising the oligo primer analysis software Oligo7 ( A single gene was chosen from each pathogen to be the target of PCR amplification. The primer and probe design principal for the multiplex real-time RT-PCR assay followed the pattern of selecting conserved gene sequence regions. Validating tests were conducted including testing of specificity using: Basic Local Alignment Search Tool (BLAST,, the evaluation of the hairpin of internal primers, primer-dimer potential, G-C content and the melting temperatures of the primers and probes by Oligo7, and sequence comparison analysis by Bioedit software ( The expected amplicon sizes were between 83 and 127 bp. Primer melting temperature ranged from 55 to 65 °C. TaqMan probes used to determine subgroup specific amplification in each reaction were individually labelled diverse report dyes including FAM, Texas Red, CY5, and VIC on the 5′ end and quencher dyes MGB and BHQ2 on the 3′ end, to differentiate the different pathogens in the same reaction system. All primers and probes were synthesized by Da’an Gene (Guangzhou, China).

Table 6 Primers and probes used in the multiplex real-time RT-PCR.

Establishment of the multiplex real-time RT-PCR assay

A three-tube, four-plex, one-step RT-PCR system was developed to accommodate the simultaneous detection of the selected nine pathogens. The pathogen subgroups were designed as follows: tube A: IVA, IVB, and RSV; tube B: adenovirus, Haemophilus influenzae, and Legionella pneumophila and tube C: Chlamydia pneumoniae, Mycoplasma pneumoniae, and Streptococcus pneumoniae. In the initiation stage, the nine sets of primer–probe pairs targeting the selected pathogens were validated using single plex PCR reactions to confirm the optimized conditions (annealing time/temperature, primer, and probe concentrations). In addition to the pathogen specific sets of primer–probe pairs, each tube of the three-tube system contained a primer–probe pair for GAPDH, to serve as an internal reference gene. The multiplex real-time PCRs were performed in 25 µl of a reaction mixture which comprised of: primer concentrations ranging from 2 to 10 pmol, fluorescent probe concentrations ranging from 5 to 10 pmol, a 6 U hot start Taq enzyme, 20 U RNasin (supplied by Da’an Gene, China), 10 U MMLV (Fapon Biotech, Shenzhen, China), 75 nmol Mg2+, 1.5 µmol Tris–HCl (pH 8.8), 0.25 µmol (NH4)2SO4, 1.25 µmol KCl, (all from Sigma-Aldrich, Germany), 0.875 µmol deoxynucleotide triphosphates (A:C:G:T = 1:1:1:1, supplied by Promega, USA), 5 µl template DNA or RNA and diethyl pyrocarbonate-treated water. An ABI 7500 Real-Time PCR system (Thermo Fisher Scientific, USA) was used for the experiment. RT was performed at 50 °C for 15 min, followed by 15 min at 95 °C, this was followed by 45 cycles of 15 s at 94 °C and 45 s at 55 °C, which amplified the cDNA and DNA. Fluorescent signals were recorded during the annealing phase of the 45 cycles. Cycle threshold (Ct) values were determined by the 7500 Real-Time PCR software at the automatic threshold setting with the Ct value of 38 chosen as a cut-off value for defining positivity. Each multiplex PCR run consisted of one negative control and one positive control. The positive control consisted of a mixture of nucleic acids of known concentration extracted from the reference viral and bacterial strains. All PCR-relevant procedures were carried out in designated PCR suites, with separate processes in different rooms, running routine decontamination. The total experimental turnaround time, from the initiation of nucleic acid extraction to the completion of target genes fluorescent signal production, was around 3.0 h.

The absolute quantification of nucleic acids by droplet digital PCR

A droplet digital PCR (ddPCR) technique was utilised to directly and precisely quantify the copy numbers of nucleic acids extracted from viral and bacterial cultures. Herein, a One-Step ddPCR Advanced kit for Probes (Bio-Rad Laboratories, Hercules, CA) was used in accordance with the manufacturer’s recommendation and as previously described36. Briefly, the reaction mixtures were assembled with 5 µl ddPCR supermix, 40 U Reverse Transcriptase, 1 µl 300 mM Dithiothreitol (DTT), the same TaqMan primers and probes used in the multiplex real-time RT-PCR (final concentration of 500 and 250 nM, respectively), and 5 µl template nucleic acids in a final volume of 20 µl. Each reaction was loaded into the sample well of an eight-well droplet cartridge together with 70 µl of droplet generation oil (Bio-Rad). Following their formation in a QX200 droplet generator (Bio-Rad), droplets were then transferred to a 96-well PCR plate, which was heat-sealed with foil before, amplification was performed using a C1000 Touch Thermal Cycler (Bio-Rad) with the following RT-PCR parameters: a reverse transcription step at 50 °C for 60 min; initial denaturation at 95 °C for 10 min, followed by 45 cycles of 94 °C for 30 s and 55 °C for 1 min; and a final extension step at 98 °C for 10 min. The PCR plate was subsequently scanned on a QX200 droplet reader (Bio-Rad) and the copies/µl of each queried target per well were analysed with QuantaSoft software version 1.7 (Bio-Rad). Droplet positivity was determined by fluorescence intensity and only droplets above a minimum threshold of fluorescence amplitude were judged as positive. By utilising ddPCR, the concentrations in copy/ml of the reference viral and bacterial strains used for the establishment of standard curves were determined. All experiments were performed in triplicate.

The evaluation of sensitivity and specificity

Twenty-five strains, namely: H1N1, H1N1 (2009), H3N2, H5N1, H7N9, IVB Yamagata, IVB Victoria, RSV (types A and B), Haemophilus influenzae, Legionella pneumoniae (subsp. fraseri Brenner et al. and subsp. pneumophila Brenner et al.), Streptococcus pneumoniae (Serotype 19F and Klein Chester), Mycoplasma pneumoniae, Chlamydia pneumoniae, and adenovirus (types 1, 2, 3, 4, 5, 7, 46, 48 and 55), were used to determine the analytical sensitivity of our assay. Serial ten-fold dilutions of the 25-target pathogenic DNA/RNA samples with known starting concentrations determined by ddPCR were subject to detection by the multiplex TaqMan RT-PCR assay, to determine the assay linearity and limits of detection. The Ct values obtained from the serial dilutions were graphed on the Y axis against the log of the dilution on the X axis, to generate standard curves for each target gene. The slope of each standard curve was calculated in accordance with the equation: E = 10(−1/slope) − 1, to determine the efficiency (E). The additional viral and bacterial species unrelated to the target pathogens, used to evaluate the specificity of the multiplex TaqMan RT-PCR assay, are shown in Table 7. A mixture of undiluted nucleic acid from each of these microorganisms was simultaneously tested alongside the target nucleic acid to assess the absence of non-specific amplification.

Table 7 Pathogenic microorganisms, including culture strains and laboratory isolates, used as the control pathogens in evaluation of the three-tube multiplex real-time RT-PCR.

The validation of reproducibility

Intra-assay reproducibility tests were carried out in triplicate by respectively testing three different concentration mixtures (105 and 103 copies/ml and a lowest limit of detection, either 500 or 250 copies/ml) of each pathogenic nucleic acid within the same experiment. The inter-assay variability was examined by repeating the intra-assay run on three continuous days to validate the reproducibility of the multiplex real-time RT-PCR.

Clinical sample verification

Nucleic acids extracted from aliquots of the 179 clinical nasopharyngeal/throat swab specimens were tested using the multiplex real-time RT-PCR. The viral results were compared with conventional immunofluorescence assay results that employed a D3 Ultra DFA Respiratory Virus Screen and ID kit. These kits used fluorescein-conjugated monoclonal antibodies that targeted the viral antigens of IVA, IVB, parainfluenza virus types 1, 2 and 3, RSV and adenovirus (Supplied by Diagnostic Hybrids, USA) (performed by the central laboratory of the collaborative hospital). Nucleic acids from all 179 samples were further analysed by Sanger sequencing which was designed as the reference method in this study. The samples in which the PCR data was inconsistent with the immunofluorescence results were confirmed by gene sequencing. Furthermore, the possibility that the pathogenic microorganisms carried in the clinical specimens belonged to five other pathogens (Haemophilus influenzae, Legionella pneumophila, Chlamydia pneumoniae, Mycoplasma pneumoniae and Streptococcus pneumoniae) was validated by gene sequencing.

Nucleotide sequencing

Automated sequencing of the nine genes belonging to the studied pathogens was carried out at Sangon Biotech (Shanghai, China) using an Applied Biosystems 3730xl DNA Analyzer (Thermo Fisher Scientific, Foster City, USA), following the contractor’s procedure. Briefly, preliminary amplification of the nucleic acids from the nine pathogens was completed in our laboratory by conventional RT-PCR and the products were sent for DNA sequencing. The sequences of PCR primers for sequencing are listed in Table 8. The sequencing results were compared with the sequences in GenBank using the BLAST algorithm.

Table 8 Sequences of PCR primers for DNA sequencing.

Statistical analysis

Consistency between the results of immunoassays, sequencing and real-time PCR assays was verified by applying Cohen’s kappa test using SPSS software, version 21.0 (IBM Corp., Armonk, NY, USA). The kappa coefficient (95% CI) value was graded as follows: 0–0.20, small; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial and 0.81–1, near perfect agreement. A p value of < 0.05 was considered statistically significant.

#Development #diagnostic #assay #threetube #multiplex #realtime #PCR #simultaneous #detection #microorganisms #causing #acute #respiratory #infections #Scientific #Reports

Leave a Comment

Your email address will not be published.