A triplex quantitative reverse transcription PCR (RT-qPCR) was designed for simultaneous detection of NoV-GI, NoV-GII and SARS-CoV-2. The primer and TaqMan probe sequences for NoV-GI and NoV-GII referred to previously published oligonucleotide primers (Liu et al., 2020). In order to cover as many NoV mutants as possible, three primers were designed to amplify NoV-GI and NoV-GII, and the TaqMan probes targeted ORF1-ORF2 junction region. In the nucleic acids detection of SARS-CoV-2, the N gene usually exhibits superior sensitivity and specificity compared with ORF1a/b target site (Zoka and Beko, 2020; Abbasi et al., 2022). Therefore, the N gene was used as the only target for SARS-CoV-2 detection. The primers and probes targeting the N gene of SARS-CoV-2 were consistent with the sequences provided by the National Microbiology Data Center (NMDC) of China, as described in https://nmdc.cn/nCoV. Primers, TaqMan probes and the target sites were described in Table 1. All the primers and probes were synthesized by Shanghai Sangon Biological Engineering Co., Ltd. (Shanghai, China) and purified by high performance liquid chromatography (HPLC).

Nucleic acid extraction kit (magnetic beads method, Guangzhou Daan Gene Co., Ltd. China) and nucleic acid extractor were used to extract nucleic acids from clinical oropharyngeal swabs and fecal specimens. Viral RNA extraction was performed following the standard protocol (Chen et al., 2020). Oropharyngeal swab specimens can be directly used to extract nucleic acids, while fecal specimens need to be dissolved with an appropriate amount of physiological saline and shaken well, and 200 μL of supernatant after centrifugation was taken for nucleic acid extraction. The viral RNA was finally eluted with ~50 μL elution buffer and subsequently frozen in −80°C refrigerator.

The RT-qPCR assays were performed using TaqMan probe one-step RT-qPCR 5G preMix (TOROIVD, China). 12.5 μL 2 × 5G qPCR buffer and 1.3 μL Enzyme Mix were added to the 25 μL reaction system. The final concentrations of primers in RT-qPCR assay were 400 nM, and the concentrations of TaqMan probes were 160 nM. Meanwhile, 5 μL template was added, and supplemented with ddH₂O to 25 μL. In the triplex RT-qPCR assay, there are three pairs of amplification primers and probes, which can simultaneously detect viral RNA of NoV-GI, NoV-GII and SARS-CoV-2.

The reverse-transcription step was performed at 50°C for 15 min followed by incubation at 95°C for 5 min. Amplification included 45 cycles of denaturation at 95°C for 10 s, followed by annealing, extension and data acquisition at 60°C for 40 s on Applied Biosystems 7500 Fast Real-Time PCR system (Thermo fisher).

The high concentration of SARS-CoV-2 RNA was used as the standard for N gene detection, and the virus RNA was extracted from an oropharyngeal swab sample of a SARS-CoV-2 strongly positive patient (cycle threshold (CT) value was 20.2). The concentration of the positive standard (1.1 × 10⁶ copies/mL) was determined by using a serial 10-fold dilution of the SARS-CoV-2 pseudovirus BDS-S1 (BDS Biological Technology Co., Ltd., Guangzhou, China; Cat. No. BDS-BW-118) as a quantitative standard curve for qPCR. The positive standards for NoV-GI and NoV-GII were extracted from RNA pseudovirus containing NoV gene fragments (Landes Medical Technology Co., Ltd., Hubei, China; Cat. No. FX10132303002). According to the results provided in the instructions, the concentrations of NoV-GI and NoV-GII standards were 2.2 × 10⁶ and 3.6 × 10⁶ copies/mL, respectively.

The prepared three viral RNA standards (NoV-GI, NoV-GII and SARS-CoV-2) were mixed equally, and serially diluted 10-fold with ddH₂O to five dilutions (1×, 10×, 100×, 1,000×, 10,000×) for each target. The triplex and singleplex RT-qPCR assays were used to detect these viral RNA standard dilutions. Cycle threshold (Ct) values were plotted against log10 viral RNA concentrations to establish a standard curve, and linear regression analysis was performed for the three targets, allowing determination of the correlation coefficient (R²). The amplification efficiencies (E) of the reactions were calculated from the curves using the equation E = (10^(−1/slope) − 1) × 100% (Ni et al., 2021).

The three viral RNA standards (NoV-GI, NoV-GII and SARS-CoV-2) were prepared and mixed in equal proportions. The mixed standards were then diluted 10-fold and 1,000-fold with ddH₂O as high and low concentration templates, respectively. The singleplex and triplex RT-qPCR assays were employed to detect the target sites of NoV-GI, NoV-GII and SARS-CoV-2 in 12 parallel experiments. To assess the reproducibility of RT-qPCR assays, coefficient of variation (CV) for Ct values were compared.

To investigate the detection specificity of the triplex RT-qPCR assay, we collected positive controls for nucleic acids detection of gastrointestinal and respiratory virus, such as rotavirus, adenovirus, influenza A virus, influenza B virus, respiratory syncytial virus and human rhinovirus. These controls were pseudoviruses that contain specific viral fragments and derived from commercialized detection kits (Guangzhou Daan Gene Co., Ltd. China). The specificity is characterized by the interference of other viruses on detection results of the triplex RT-qPCR assay.

The 95% lower limit of detection (LLOD95) and lower limit of quantification (LLOQ) of RT-qPCR were assessed by using the viral RNA positive standard as described above. LLOD95 was defined as the target concentration (dilution level) that could be detected with 95% probability. The positive standards of NoV-GI, NoV-GII and SARS-CoV-2 were diluted serially in sterile nuclease-free water. Eight concentrations ranging from 1: 400, 1: 2,000, 1: 4,000, 1: 10,000, 1: 20,000, 1: 40,000, 1: 80,000 and 1: 200,000 were tested in parallel by the target specific singleplex and triplex RT-qPCR assay to compare the sensitivity of detection. The RT-qPCR of each dilution was performed in quadruplicate, and the RT-qPCR was independently repeated four times. A total of 16 tests were performed for each concentration, and positive detection rates were counted at each dilution. LLOD is the measurand concentration that produces at least 95% positive replicates, and the LLOD95 was estimated by using logistic regression with target concentration (dilution level) as an explanatory variable and detection of the sample (pos/neg) as a response variable. The logistic regression curve is obtained by plotting:

Where x denotes log2 (concentration), and the two unknown parameters β₀ and β₁ are approximated by maximum likelihood (ML) estimation (Forootan et al., 2017; Persson et al., 2018).

The LLOQ was calculated based on the reported method (Forootan et al., 2017). This was done by calculating the SD for the responses of the replicate samples at the different concentrations, and the SD of the data was calculated from corresponding template concentration. The coefficient of variation (CV = 100 × SD/mean) for concentrations measured in replicates with qPCR were plotted to Log10 (measured template concentration).

Clinical performance of the triplex RT-qPCR assay was evaluated by detecting different clinical specimens. Oropharyngeal swabs and fecal specimens of patients in Jingzhou First People’s Hospital (Jingzhou, China) from January to March 2023 were collected. A total of 351 oropharyngeal swabs were collected for detection the nucleic acid of SARS-CoV-2, and 32 positive specimens were detected by the commercial kit of Wuhan EasyDiagnosis Biomedicine Co., Ltd. (Hubei, China). One hundred and eighty-eight fecal specimens for NoV detection were collected, and 34 positive samples were detected by commercial kit of Landes Medical Technology Co., Ltd. (Hubei, China). The newly designed singleplex and triplex RT-qPCR assays were also used to detect these samples, and Bland–Altman agreement analysis was used to highlight the correlation between the Ct values of the singleplex and triplex assays using GraphPad Prism (GraphPad Software Inc., La Jolla, CA, United States).

In order to investigate the detection range of the triplex RT-qPCR, the viral RNA standards of NoV-GI, NoV-GII and SARS-CoV-2 were serially diluted 10-fold and serve as templates for detection. Thus, the amplification plots for triplex and singleplex RT-qPCR assays were obtained (Figures 1A–C).

Cycle threshold (Ct) values were plotted against log10 viral RNA concentrations to establish a standard curve, and there was no significant difference between triplex and singleplex assay curves (Figures 1D–F). The R², slopes and amplification efficiency (E) of the standard curves were marked on Figure 1. The R² of the standard curves were greater than 0.99. The E values of each detection target were within the range of 102–110%, and the results were in line with requirements (Broeders et al., 2014).

To explore the reproducibility of RT-qPCR, we used singleplex and triplex RT-qPCR assays to detect nucleic acids of NoV and SARS-CoV-2 at high and low template concentrations, respectively. A total of 12 parallel experiments were conducted, the average Ct value and coefficient of variation (CV) of each target were shown in Table 2. The Ct values of the singleplex and triplex RT-qPCR assays did not differ significantly (p > 0.05). Meanwhile, the CV of Ct values were <1%. The results indicate that triplex RT-qPCR assay has high repeatability and is suitable for clinical detection.

The triplex and singleplex RT-qPCR assays were tested on nucleic acids extracted from other gastroenteritis and respiratory viruses, including rotavirus, adenovirus, influenza A virus, influenza B virus, respiratory syncytial virus and human rhinovirus. The nucleic acids of these viruses were used as templates and then RT-qPCR experiments were performed. Neither assay found non-specific amplification signals for any target nucleic acids from these tested organisms, suggesting that the assays have good specificity.

The LLOD95 was utilized to evaluate the sensitivity of the RT-qPCR assays. The viral nucleic acid standards were serially diluted, and then the positive detection rates of three targets NoV-GI, NoV-GII and SARS-CoV-2 at each dilution were detected, as shown in Figures 2A–C. LLOD95 was estimated by logistic regression analysis as previously described (Persson et al., 2018). The LLOD95 results of triplex and singleplex RT-qPCR were shown in Figure 2D.

For the singleplex RT-qPCR assay, the LLOD was calculated to be 125.4 copies/mL for NoV-GI, 110.3 copies/mL for NoV-GII, and 142.0 copies/mL for SARS-CoV-2 (Figure 2D). For the triplex RT-qPCR assay, the LLOD was calculated to be 160.7 copies/mL for NoV-GI, 128.5 copies/mL for NoV-GII, and 172.8 copies/mL for SARS-CoV-2. The LLOD values of triplex assay were slightly lower than that of the singleplex. Additionally, there was no remarkable difference in LLOD values between them. This result indicates that the triplex and singleplex assays have the semblable sensitivity. In addition, lower limit of quantification (LLOQ) can also be estimated from the replicate standard curves (Forootan et al., 2017). The LLOQ of singleplex and triplex RT-qPCR assay for each target site were 309.7–393.6 copies/mL and 378.4–494.3 copies/mL, respectively (Supplementary Figure S1), and there was no significant difference in LLOQ between the two RT-qPCR assays.

A total of 351 oropharyngeal swabs for detection of SARS-CoV-2 nucleic acid, and 188 fecal specimens for NoV detection were collected. Meanwhile, the nucleic acids of these samples were extracted using the magnetic beads method. The newly designed singleplex and triplex RT-qPCR detection assays were used to assess these samples. A total of 32 positive samples of SARS-CoV-2 and 34 positive samples of NoV were detected. The number of positive samples was consistent with the commercial detection kit. Interestingly, all 34 positive specimens of NoV were NoV-GII genogroup, which maybe result from the fact that NoV-GII is the most predominant NoV genogroup that causes human diseases in China (Kobayashi et al., 2016; Yu et al., 2022). In addition, one case positive for SARS-CoV-2 was detected in 34 NoV-positive fecal specimens.

The distribution frequencies of Ct values for NoV-GII and SARS-CoV-2 of the clinical samples were displayed in Figure 3. For triplex and singleplex RT-qPCR, there was no significant difference in the proportion of each Ct value range. The Bland–Altman agreement analysis was used to highlight the correlation between the Ct values, and the agreement analysis plot was shown in Figure 4. 91.2% (31/34) of the positive points for the NoV-GII were within the 95% limits of agreement (LoA), and 96.8% (31/32) of the positive points for the SARS-CoV-2 were between the 95% LoA. The mean of Ct ratio for NoV-GII and SARS-CoV-2 was close to 1. This result indicates that the triplex and singleplex RT-qPCR assays have great consistency for detecting clinical specimens.