key: cord-0261662-ezu5k23k authors: Struijk, R.; van den Ouden, A.; McNally, B.; de Groot, T.; Mulder, B.; de Vos, G. title: Ultrafast RNA extraction-free SARS-CoV-2 detection by direct RT-PCR using a rapid thermal cycling approach date: 2021-11-11 journal: nan DOI: 10.1101/2021.11.09.21265517 sha: afc3a719e2d56e62bdf5cad4c765b418d0d70217 doc_id: 261662 cord_uid: ezu5k23k The surging COVID19 pandemic has underlined the need for quick, sensitive, and high-throughput SARS-CoV-2 detection assays. Although many different methods to detect SARS-CoV-2 particles in clinical material have been developed, none of these assays are successful in combining all three of the above characteristics into a single, easy-to-use method that is suitable for large-scale use. Here we report the development of a direct RT-PCR SARS-CoV-2 detection method that can reliably detect minute quantities of SARS-CoV-2 gRNA in nasopharyngeal swab samples as well as the presence of human genomic DNA. An extraction-less validation protocol was carried out to determine performance characteristics of the assay in both synthetic SARS-CoV-2 RNA as well as clinical specimens. Feasibility of the assay and analytical sensitivity was first determined by testing a dilution series of synthetic SARS-CoV-2 RNA in two different solvents (water and AMIES VTM), revealing a high degree of linearity and robustness in fluorescence readouts. Following analytical performance using synthetic RNA, the limit of detection was determined at equal to or less than 1 SARS-CoV-2 copy/ul of sample in a commercially available sample panel that contains surrogate clinical samples with varying SARS-CoV-2 viral load. Lastly, we benchmarked our method against a reference qPCR method by testing 87 nasopharyngeal swab samples. The direct endpoint ultra-fast RT-PCR method exhibited a positive percent agreement score of 98.5% and a negative percent agreement score of 100% as compared to the reference method, while RT-PCR cycling was completed in 27 minutes/sample as opposed to 60 minutes/sample in the reference qPCR method. In summary, we describe a rapid direct RT-PCR method to detect SARS-CoV-2 material in clinical specimens which can be completed in significantly less time as compared to conventional RT-PCR methods, making it an attractive option for large-scale SARS-CoV-2 screening applications. At the end of 2019, a novel strain of betacoronavirus called 2019_nCoV was identified 1 in Wuhan, 34 Hubei Province, China. In the following months, the virus proved highly contagious and as of 35 September 20 th , 2021, a total of 228,394,572 confirmed cases of COVID-19 including 4,690,186 deaths 36 have been reported according to the World Health Organization Coronavirus (COVID-19) Dashboard. 37 Shortly after its identification, 2019_nCoV was renamed to the current consensus 2 term SARS-CoV-2 38 and the first complete SARS-CoV-2 RNA sequence was made public by the Chinese Centers for 39 Disease Control and Prevention. 40 The emergence of SARS-CoV-2 has underlined the need for rapid and easy-to-implement diagnostic 41 nucleic acid detection methods. Nucleic acid amplifications tests (NAAT) such as multiplex reverse- is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Analytical Q Panel 01 (cat. no. SCV2AQP01-A, Qnostics Ltd, United Kingdom) described in Table 2 . Wilhelmina Hospital (Nijmegen, The Netherlands) using ∑-Transwab® nasopharyngeal swabs (Copan 78 Diagnostics, Italy). After collection, swabs were directly transferred to a sterile microtube containing 79 1mL AMIES liquid and heat-inactivated at 100°C for 10 minutes. An aliquot was taken from the tube 80 containing the swab and subjected to RNA extraction followed by qPCR measurements (reference 81 method), and the remaining volume was shipped to our laboratory in Goes where it was subjected to 82 direct RT-PCR as described below approximately 24-48h after primary sampling. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint gRNA and human gDNA material in a sample. In the reaction, FAM-labelled (fluorescein amidites) 89 oligonucleotide probes are used to detect SARS-CoV-2 N1 and ORF1ab gene sequences and Cy5-90 labeled oligonucleotide probes are used to detect a sequence in the human RPP30 gene. Primers/probes 91 specific for RPP30 were included as an internal control target, serving both to confirm absence of PCR 92 inhibition and indicate the presence of human genomic material in sample wells, aiding laboratory 93 personnel to verify that sampling and pipetting went correctly. Primer and probe oligonucleotide 94 sequences were designed using extensive bioinformatical analyses and in silico digital PCR to minimize 95 amplification of off-target sequences and formation of primer-dimers or other secondary structures. 96 Cross-reactivity of the reagents with other respiratory viruses was evaluated and showed no 97 amplification of targets in off-target organism genomic sequences (data not shown). A total of 16μl of 98 RT-PCR master mix containing 10μl RT-PCR Chemistry 2x, 1.6μl SARS-CoV-2/hRNaseP Primers and 99 Probes and 4.4μl nuclease-free H2O (MBS, Netherlands) was added to 4μl of sample resulting in a total 100 reaction volume of 20µl. The microplate was heat-sealed after pipetting with a transparent Clear Heat 101 Seal (MBS, Netherlands) on a NextGenPCR Semiautomatic Heat Sealer (MBS, Netherlands) and 102 transferred to a NextGenPCR machine for thermal cycling using the RT-PCR program detailed in Table 103 3. Three wells containing Human Positive Control material from the MBS detection assay were 104 included to confirm efficient PCR cycling. After PCR was completed, the sealed microplate was 105 snapped to an imaging anvil (MBS, Netherlands), transferred to a FLUOstar Omega Microplate Reader 106 (BMG Labtech, Germany), scanned and fluorescence readout results exported to Excel for downstream 107 data analysis. For measurements in synthetic RNA, the microplate was also scanned on a Bio-1000F 108 Gel Imager (Microtek, Taiwan) and the results interpreted using custom designed QuickDetect software 109 (MBS, Netherlands). 110 The SARS-CoV-2 qPCR detection method described by Corman and colleagues 4 was used as the 112 reference method for this study. In short, nucleic acids were extracted from the sample/AMIES mixture is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint robotic workstations (Roche Diagnostics GmbH, Mannheim, Germany), an internal control sequence 116 specific for phocine distemper virus (PhDV) was added to the sample prior to nucleic acid extraction. 117 After extraction, 5 μl TaqMan Fast Virus 1-Step Master Mix (ThermoFisher Scientific) and 5 μl primers 118 and probes (Table 4 ) were added to 10 μl of spiked-in sample. Thermal cycling was performed on a 119 LC480-II instrument (Roche) using the RT-PCR program detailed in Table 3 . Data analysis was 120 performed using FLOW software (Roche) and a threshold Ct-value of 40 was used as cut-off to interpret 121 results as positive (Ct ≤ 40), negative (no Ct after 50 cycles) or indeterminate (Ct between 40-50). 122 Direct RT-PCR on synthetic RNA dissolved in water or AMIES liquid shows similar 124 amplification linearity and limit of detection 125 We first determined the analytical sensitivity of our assay in a serial logarithmic dilution series of 126 synthetic RNA (Table 1) . Samples were diluted in nuclease-free water to establish a baseline reading, 127 as well as AMIES liquid to detect possible discrepancies of the fluorescence readouts using an actual 128 VTM. A sample was called as being detected when all three replicates showed a significantly higher 129 fluorescence level (sample RFU ≥ mean RFU in NTC + 3 × standard deviation) as compared to 130 background signal in non-template control (NTC) wells within the same medium. 131 Using the above cut-offs, FAM fluorescence was detected on a Bio-1000F blue light scanner in 132 synthetic RNA dilutions ranging from 1 to 10,000 SARS-CoV-2 copies/µl in both solvents, but not in 133 the 0.1 copies/µl sample ( Figure 1A) . Dilutions made in nuclease-free H2O as compared to AMIES 134 liquid were shown to have slightly higher R 2 -values (R 2 = 0.91 and R 2 = 0.87 in nuclease-free water 135 and AMIES liquid, respectively) as well as a higher signal amplitude when measured on a FLUOstar 136 Omega microplate reader, most noticeable in samples with a SARS-CoV-2 concentration ≥ 100 137 copies/µl ( Figure 1B) . The FLUOstar Omega is equipped with multiple single channel fluorescent 138 detectors allowing for qualification of FAM signals as well as CY5 signals, the latter of which cannot 139 be detected with a Bio-1000F blue light scanner. Human RPP30 was detected in the kit positive control 140 only (p < 0.05, two-sided Student's t-test), indicating that the used primer oligonucleotides are specific 141 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint for the intended RPP30 human genomic target without off-target amplification in samples lacking 142 RPP30 templates ( Figure 1C) . 143 144 SARS-CoV-2 detection without prior nucleic acid extraction 145 Next, a commercially available analytical sample panel containing samples with a known concentration 146 of inactivated SARS-Cov-2 particles (ranging from 1 x 10 6 to 5 x 10 1 copies per milliliter of sample) 147 mixed with a human cell matrix was tested using the direct RT-PCR protocol ( Table 2) is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint were positive for SARS-CoV-2 and were likely undetected due to competition for PCR reagents 168 between primer sets, a phenomenon that is well described in multiplex PCR which we could also 169 observe here (see Figure 2 , reduced CY5 label fluorescence in samples with high SARS-CoV-2 copy 170 number). We observed strong correlation between fluorescence readouts following direct RT-PCR and 171 threshold cycle (Ct) numbers with a correlation coefficient of R 2 = 0.84 and concordance statistic of ρc 172 = 0.79 between the two techniques (Figure 3) . A clinical agreement study was carried out to estimate 173 percent positive agreement (PPA, or clinical sensitivity) and percent negative agreement (NPA, or 174 clinical specificity) scores as recommended by the FDA in the "CLSI EP12: User Protocol for 175 Evaluation of Qualitative Test Performance" protocol. As detailed in Figure 4B , PPA was calculated 176 at 98.5% and NPA was calculated at 100%. CoV-2 (see Table 5 ) were analyzed on a BioRad CFX96 qPCR system in addition to NextGenPCR 182 analysis as described above. No off-target amplification of FAM-labeled fragments was observed in 183 these samples (CT undetermined on CFX96, NextGenPCR RFU ranging from ~16,000 to 20,000 in the 184 FAM channel with a signal of 18,000 RFU in the negative SPEC0004 sample), except for three samples 185 where 400 copies/µl of MBS SARS-CoV-2/RPP30 positive control material was spiked-in prior to 186 sample workup to control for potential PCR inhibition due to buffer components (Figure 5) . The signal 187 amplitude in the spike-in samples averaged to 142,707 RFU in the FAM channel, well above the average 188 is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint without prior nucleic acid extraction 5, 6 . In this paper, we describe the use of the NextGenPCR SARS-194 CoV-2 detection chemistry and NextGenPCR thermal cycler to accurately detect SARS-CoV-2 RNA 195 in clinical samples with significantly reduced PCR cycling time and without the need for sample lysis 196 or nucleic acid extraction. The analytical sensitivity of our developed assay was determined at 1.0 × 10 0 197 copies/ul. The same limit of detection was observed using direct RT-PCR in the Qnostics sample panel, 198 confirming that clinical samples with a SARS-CoV-2 viral load of ≥ 1 copy/ul can be consistently 199 detected in our assay. When compared to a reference quantitative RT-PCR method targeting the SARS- was confirmed by a second study 9 that compared five RT-PCR master mixes for SARS-CoV-2 detection 219 by multiplex RT-PCR describing improved detection rates of N1 and N2 target sequences after thermal 220 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint sample lysis at 98°C (81%) as compared to lysis at 65°C (56%) or no thermal lysis (52%) prior to RT-221 PCR testing. 222 Apart from SARS-CoV-2 testing, the NextGenPCR thermal cycling system is also suitable for the is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint Target concentration (dC/ml) SCV2AQP01-S01 1,000,000 SCV2AQP01-S02 100,000 SCV2AQP01-S03 10,000 SCV2AQP01-S04 5,000 SCV2AQP01-S05 1,000 SCV2AQP01-S06 500 SCV2AQP01-S07 100 SCV2AQP01-S08 50 SCV2AQP01-S09 0 305 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 11, 2021. ; https://doi.org/10.1101/2021.11.09.21265517 doi: medRxiv preprint A Novel Coronavirus from Patients with Pneumonia in China Coronaviridae Study Group of the International Committee on Taxonomy of, V. The species 243 Severe acute respiratory syndrome-related coronavirus: classifying 2019-nCoV and naming it 244 SARS-CoV-2 Molecular detection of respiratory viruses Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR We would like to thank Martin Donker (Isogen Life Sciences B.V., Netherlands) for kindly providing 238 an aliquot of the SARS-CoV-2 Analytical Q Panel 01. 239