Background
The global coronavirus disease 2019 (COVID-19) pandemic, caused by infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is posing an enormous burden on social, economic, and healthcare systems worldwide [
1]. As there is currently no specific treatment option, early detection of SARS-CoV-2-infected patients has facilitated effective isolation and treatment to prevent disease spread. Currently, clinical diagnosis of COVID-19 is mainly confirmed by detecting SARS-CoV-2 RNA using reverse transcriptase real-time quantitative polymerase chain reaction (qRT-PCR) [
2‐
5]. However, the sensitivity and reliability of qRT-PCR has been questioned due to cases of negative results in some patients who were highly suspected of having the disease based on clinical presentation and exposure history [
6,
7].
Droplet digital PCR (ddPCR) is a third generation PCR based on the principles of limited dilution and Poisson statistics [
8,
9], which works by separating a sample into thousands to millions of droplets and then partitioning them to be read as either positive or negative depending on fluorescence amplitude [
10‐
13]. These vast and highly consistent oil droplets substantially improve the detection dynamic range and accuracy of ddPCR [
14]. In recent years, ddPCR has found many applications, such as analysis of viral load from clinical samples, detection of rare mutations, analysis of copy number variation, and precise miRNA quantification [
15‐
17].
Given the high sensitivity of ddPCR, Zhao JK et al
. utilized this technique to evaluate the viral loads of SARS-CoV-2 from upper respiratory tract specimens for the first time, showing that ddPCR can accurately reflect the viral loads of such specimens, especially nasopharyngeal swabs [
18]. Subsequently, Lu et al
. [
19] used serial dilutions of the same clinical samples to demonstrate that the LoD (limit of detection) of ddPCR is at least 10 times better than that of qRT-PCR. However, the limitation of the ddPCR assay is that it often needs unique supporting reagents, instruments, and professional operators, causing high running costs with moderate throughput. More convenient and sensitive methods are urgently needed as alternative diagnostic approaches for detecting SARS-CoV-2.
Nested PCR typically utilizes two sequential amplification reactions, each of which uses a different pair of primers, resulting in an increase in sensitivity and specificity. The product of the first amplification reaction is used as the template for the second, which is primed by oligonucleotides that are placed internal to the first primer pair. The use of two pairs of oligonucleotides allows for a higher number of cycles to be performed, thereby increasing the sensitivity of the PCR. Feng et al. [
20] previously developed a novel locked nucleic acid (LNA)-based one-step single-tube nested (OSN)-qRT-PCR strategy to detect viral and bacterial pathogens with higher sensitivity and specificity than qRT-PCR and without the need of lid opening. To improve the diagnostic accuracy of nucleic acid detection of SARS-CoV-2 in low viral load samples, they developed and evaluated the sensitivity and accuracy of OSN-qRT-PCR in detecting SARS-CoV-2 [
21]. However, the analytical performance of OSN-qRT-PCR in the published study is insufficient, lacking information such as specificity, reportable range, and the LoD. In addition, no studies have been conducted comparing the clinical application value of OSN-qRT-PCR and ddPCR for detecting SARS-CoV-2.
Here we provide a comparison of OSN-qRT-PCR and ddPCR with qRT-PCR using a dilution series of SARS-CoV-2 pseudoviral RNA and clinical samples. The detectable range and sensitivity of each assay were determined and clinical samples (n = 34) were used to validate clinical sensitivity and specificity. Compared with qRT-PCR and ddPCR, OSN-qRT-PCR showed higher sensitivity and greater practicality, making it better suited for detecting SARS-CoV-2 in low viral load samples.
Materials and methods
Ethics statement
The Ethics Committee of The Second People’s Hospital of Fuyang approved this study. Existing samples collected during standard diagnostic tests were tested and analyzed by qRT-PCR, OSN-qRT-PCR, and ddPCR. No extra burden was posed to patients.
Specimen collection
We retrospectively identified 24 hospitalized patients clinically diagnosed with COVID-19 between January 30, 2020, and February 17, 2020, in The Second People’s Hospital of Fuyang. Throat (n = 18) and anal (n = 4) swabs, sputum (n = 10), and blood (n = 2) samples were collected from the enrolled patients. All aspects of the study were performed according to national ethics regulations and approved by the Institutional Review Boards of China Center for Disease Control and Prevention (CDC). Written consent was obtained from patients or children’s parents.
SARS-CoV-2 pseudovirus preparation
The SARS-CoV-2 reference sequence was synthesized and cloned into a lentiviral vector and pseudovirus was prepared in 293T cells. The obtained pseudovirus contained RNA sequences of the
ORF1ab and
N genes in the lentiviral genome. The SARS-CoV-2 pseudovirus used in qRT-PCR and OSN-qRT-PCR was synthesized and processed by BDS company (DA’an, Guangzhou, China) at a RNA concentration of 2.0 × 10
4 copies/ml. The SARS-CoV-2 pseudovirus used in ddPCR was synthesized and processed by BioPerfectus Technologies Co. (Taizhou, China) at a RNA concentration of 1.5 × 10
5 copies/ml. The SARS-CoV-2 pseudoviral RNA was diluted with pseudovirus diluent (dilution ratio and method is shown in Table
1), and SARS-CoV-2 pseudoviral RNA of the diluted samples S1, S2, S3, S4, S5, S6, S7, and S8 were extracted by using membrane adsorption kits (Di’an, Hangzhou, China).
Table 1
Dilution ratio and the concentration (copies/ml) of SARS-CoV-2 pseudovirus RNA standards used in each assay
S0 | 0 | 20,000 | 0 | 150,000 |
S1 | 5 | 4000 | 5 | 30,000 |
S2 | 10 | 2000 | 20 | 7500 |
S3 | 20 | 1000 | 100 | 1500 |
S4 | 40 | 500 | 300 | 500 |
S5 | 100 | 200 | 600 | 250 |
S6 | 200 | 100 | 1000 | 150 |
S7 | 400 | 50 | – | – |
S8 | 1000 | 20 | – | – |
Total RNA from throat and anal swabs, sputum, and blood samples from each patient was extracted from supernatants using Reagent of Nucleic Acid Extraction or Purification (Di’an, Hangzhou, China) following the manufacturer’s instructions. SARS-CoV-2 nucleic acid detection was mainly targeted at the two-segment conserved gene sequence of its genome, located at ORF1ab and N.
ddPCR workflow
Reaction components of the ddPCR assay kit (BioPerfectus) included 5 µl of Supermix, 2 µl of reverse transcriptase, 1 µl of 300 mM DTT, 5 µl of SARS-CoV-2 reaction solution, and 7 µl template. All procedures followed the manufacturer’s instructions for the QX200 Droplet Digital PCR System using Supermix for the probe (no dUTP) (Bio-Rad, Hercules, USA). 20 µl of each reaction mix was converted to droplets with the QX200 droplet generator. Droplet-partitioned samples were then transferred to a 96-well plate, sealed, and cycled in a C1000 Touch Thermal Cycler (Bio-Rad) under the following cycling protocol: 50 °C for 60 min and 95 °C for 10 min, followed by 40 cycles of 95 °C for 30 s and 56 °C for 1 min, then 98 °C for 10 min and 4 °C hold. FAM (ORF1ab) and HEX (N) channels were selected to detect SARS-CoV-2. As digital PCR is a novel technology, R&D Company replied the primer sequence of ddPCR for detecting SARS-CoV-2 was confidential status. The cycled plate was then transferred and FAM and HEX channels read using the QX200 reader. Each run contained positive and negative controls. Samples were only considered positive when both FAM and HEX channels had signals.
OSN-qRT-PCR workflow
Reaction components of the OSN-qRT-PCR assay kit (Sansure, Changsha, China) included 20 µl of template, 26 µl of reaction buffer, and 4 µl of the enzyme mixture. After vortexing and centrifugation, the reaction tube was transferred to the LightCycler 480 II Real-Time PCR System (Roche, Basel, Switzerland). The OSN-qRT-PCR amplification reaction contained the following steps: 50 °C for 30 min, 95 °C for 1 min, 20 cycles at 95 °C for 30 s, 70 °C for 40 s, and 72 °C for 40 s, followed by 40 cycles at 95 °C for 15 s, 60 °C for 30 s, and 25 °C for 10 s of instrument cooling. FAM (
ORF1ab) and ROX (
N) channels were selected to detect SARS-CoV-2, and the VIC channel was chosen to detect the reference gene (human
ABL1). Each run contained positive and negative controls. FAM, HEX, and VIC channels all showed typical S-shaped amplification curves. OSN-qRT-PCR outer primers were designed according to the principle described in previous publications [
22,
23]. All of the outer primers for the ORF1ab gene and N gene were designed by Oligo 7 software and several locked nucleic acids (LNAs) were incorporated into outer primer nucleotides. The result was considered valid when the cycle threshold (Ct) value of the reference gene was ≤ 37. The result was considered positive when the Ct values of both target genes were ≤ 35 and negative when they were both > 35. If only one of the target genes had a Ct value ≤ 35 and the other was > 35, it was interpreted as a single-gene positive.
qRT-PCR workflow
The qRT-PCR kit (DaAn Gene; Guangzhou, China) included 17 µl of SARS-CoV-2 NC reaction solution A, 3 µl of NC reaction solution B, and 5 µl of template. After vortexing and centrifugation, the reaction tube was transferred to the LightCycler 480 II Real-Time PCR System (Roche). The primers sequences of qRT-PCR was below:
CoV-N-P: 5′FAM-TTGCCCCCAGCGCTTCA-BHQ1 3'
CoV-N-F: 5′ TTGGGGACCAGGAACTAAT 3'
CoV-N-R: 5′ GAAGGTGTGACTTCCATGC 3'
ORF1ab-P: 5′ HEX- TCCCACCCAAGAATAGCATAGATGC-BHQ1 3'
ORF1ab-F1: 5′ TTTAGATATATGAATTCACAGGGA 3'
ORF1a-R1: 5′ ACCAACACCCAACAATTTAAT 3'
RNP-F: AGATTTGGACCTGCGAG
RNP-R: ACTGAATAGCCAAGGTGAG
RNP-P: 5′Cy5- TCCACAAGTCCGCGCAGAG-BHQ2-3′
The qRT-PCR amplification reaction contained the following steps: 50 °C for 15 min, 95 °C for 15 min, 45 cycles at 94 °C for 15 s, and 55 °C for 45 s. FAM (N) and VIC (ORF1ab) channels were selected to detect SARS-CoV-2, and the CY5 channel was chosen to detect the reference gene (human ABL1). The result was considered valid when the Ct value of the reference gene was ≤ 37. The result was considered positive when the Ct values of both target genes (ORF1ab and N) were ≤ 37 and were considered negative when they were both > 40. If only one of the target genes had a Ct value fall in the gray zone (37–40), it was retested. If the repeated result was positive for only one of two targets genes, it was interpreted as positive.
Dynamic range and LoD of OSN-qRT-PCR, ddPCR, and qRT-PCR
To evaluate the dynamic range and consistency of OSN-qRT-PCR, ddPCR, and qRT-PCR, we first ran a serial dilution of the linear RNA standard for each assay. To determine the LoD, the lower concentration RNA standards (including S3–S8) were analyzed 14 times. The LoD was calculated by Probit regression analysis with a 95% repeatable probability.
Data statistical analysis
Analysis of the ddPCR data was performed with Quanta Soft Analysis Software v1.7.4 to calculate the concentration of the target. Plots of linear regression were conducted with GraphPad Prism 7.0, and Probit analysis for LoD was conducted with MedCalc software v19.2.1. Bland–Altman analysis of qRT-PCR, OSN-qRT-PCR, and ddPCR results for patient samples was evaluated by SPSS 23.0 statistical software.
Discussion
The clinical detection sensitivity of qRT-PCR is affected by various factors, such as the nucleic acid extraction method, the one-step qRT-PCR reagent used, and the primer/probe sets [
25]. It has been reported that seven commercial qRT-PCR detection kits revealed significant differences in the detection ability for weakly positive samples [
26]. ddPCR has exhibited higher sensitivity and precision than classical qRT-PCR [
27,
28]. Recent studies have confirmed that both ddPCR and OSN-qRT-PCR are strongly recommended in clinical practice for the diagnosis of COVID-19 and for follow-up of positive patients until complete remission [
21,
29]. However, ddPCR is limited to special equipment, which hinders its clinical application. Compared with ddPCR, the advantage of OSN-qRT-PCR is greater practicality because of easier adaptation for laboratories already equipped with traditional real-time PCR machines.
Here, for the first time, we provide a head-to-head comparison of OSN-qRT-PCR and ddPCR with qRT-PCR for the detection of SARS-CoV-2 using a pseudoviral RNA standard, inter-laboratory quality assessment panel, and clinical samples of different types. The detectable range, consistency, specificity, and LoD of each method were comparably analyzed. Our results demonstrate that OSN-qRT-PCR and ddPCR are reliable for quantitatively detecting SARS-CoV-2. In addition, Bland–Altman analysis showed that ddPCR and OSN-qRT-PCR had good correlation with qRT-PCR in testing clinical specimens. In particular, the detection performance of both OSN-qRT-PCR and ddPCR assays were better than qRT-PCR, and the OSN-qRT-PCR assay had the lowest LoD, suggesting that OSN-qRT-PCR and ddPCR assays are valuable additions for detecting SARS-COV-2 in samples with low viral loads. Although the sensitivity of OSN-qRT-PCR was reported to be 10-fold higher than qRT-PCR using plasmids [
21], our results revealed that the sensitivity of OSN-qRT-PCR was only 2–3-fold higher than qRT-PCR when using pseudoviral RNA.
A previous study revealed that SARS-CoV-2 exists in both the upper and lower respiratory tract and that the viral load in sputum is higher than that of throat swabs [
30]. Our findings also confirmed that although SARS-CoV-2 can colonize the upper respiratory tract, lower respiratory tract samples better reflect the viral replication level in infected patients. Although OSN-qRT-PCR and ddPCR both exhibited higher sensitivity than qRT-PCR, there were still false negative results (six missed by OSN-qRT-PCR, 11 missed by ddPCR) when analyzing clinical specimens. This may have been due to the quality of sample collection or viral loads falling below detection limits resulting from missing the optimum sample collection time. For specimen 20
#, the results of OSN-qRT-PCR and qRT-PCR were negative, while ddPCR was positive. This discordant result may have been due to the specimen type (sputum) or the various influencing factors in the nucleic acid extraction process, leading to poor stability of test results.
This study had several limitations. First, the SARS-CoV-2 pseudoviral RNA concentration used in the study for exploring detectable ranges did not include high concentrations. Second, the clinical specimens were only from COVID-19-confirmed patients in the acute phase of infection; clinical specimens from patients in the recovery phase or suspected patients were not included. Finally, our study was limited by a small sample size and thus conclusions should be interpreted with caution and confirmed by further studies.
Conclusions
We validated the implementation of OSN-qRT-PCR and ddPCR systems as new alternatives to qRT-PCR for the sensitive and accurate quantification of SARS-CoV-2, especially in samples with low viral loads. Considering its sensitivity and practicality, OSN-qRT-PCR is a highly valuable and feasible method that offers the potential to facilitate clinical diagnoses and decision-making for patients with COVID-19.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.