key: cord-0731057-h6b8f5t9
authors: Chaintoutis, S. C.; Chassalevris, T.; Tsiolas, G.; Balaska, S.; Vlatakis, I.; Mouchtaropoulou, E.; Siarkou, V. I.; Tychala, A.; Koutsioulis, D.; Skoura, L.; Argiriou, A.; Dovas, C. I.
title: A one-step real-time RT-PCR assay for simultaneous typing of SARS-CoV-2 mutations associated with the E484K and N501Y spike protein amino-acid substitutions
date: 2021-06-01
journal: nan
DOI: 10.1101/2021.05.31.21257367
sha: 2dd45acc1824c13ce97d28d0a186f97210c9e75d
doc_id: 731057
cord_uid: h6b8f5t9

The emergence of SARS-CoV-2 mutations resulting in the S protein amino-acid substitutions N501Y and E484K, which have been associated with enhanced transmissibility and immune escape, respectively, necessitates immediate actions, for which their rapid identification is crucial. For the simultaneous typing of both of these mutations of concern (MOCs), a one-step real-time RT-PCR assay employing four locked nucleic acid (LNA) modified TaqMan probes was developed. The assay is highly sensitive with a LOD of 117 copies/reaction, amplification efficiencies >94% and a linear range of over 5 log10 copies/reaction. Validation of the assay using known SARS-CoV-2-positive and negative samples from human and animals revealed its ability to correctly identify wild type strains, and strains possessing either one or both targeted amino-acid substitutions, thus comprising a useful pre-screening tool for rapid MOC identification. The basic principles of the methodology for the development of the assay are explained in order to facilitate the rapid design of similar assays able to detect emerging MOCs.

Specifically, the B.1.1.7 lineage (UK variant or VOC 202012/1; 17 amino-acid substitutions) emerged in southeast England in November 2020 and is rapidly spreading towards fixation. It has been indicated that this variant has a higher reproduction number up to 90% higher than pre-existing variants and will lead to large resurgences of COVID-19 cases (Davies et al., 2020) . The B.1.351 lineage (South African variant or S.501Y.V2; 17 amino-acid substitutions) was initially reported in South Africa in December 2020. Additionally, the P.1/B.1.1.28.1 lineage (Brazilian variant or 501Y.V3; 17 amino-acid substitutions) was reported in Brazil in January 2021 (Abdool Karim and de Oliveira, 2021) . Both of the latter have been also characterized by increased transmissibility.

In all of these VOCs, the N501Y amino-acid substitution (change of asparagine to tyrosine) is present at the position 501 of the S-protein (receptor-binding domain), which confers enhanced affinity for receptor binding (Starr et al., 2020) . The two latter VOCs (B.1.351 and P.1) also possess the E484K amino-acid substitution (change from glutamate to lysine), which reduces the neutralization sensitivity to convalescent sera (Wibmer et al., 2021) , and thus, possibly affecting the protection conferred by vaccine-derived antibodies, or antibodies produced in previous infections from non-carrying the E484K SARS-CoV-2 strains. These characteristics have a serious impact on the control of the ongoing pandemic, including the need for the modification or the development of novel diagnostic assays, or the currently available vaccines. Most importantly, besides the aforementioned VOCs, both of these substitutions (N501Y and E484K) have been reported to emerge independently in several other SARS-CoV-2 lineages worldwide, such as in lineages AP.1, A.27 for the substitution N501Y, and lineages R.2, N.10 and P.2 for the substitution E484K. As a result, rapid identification of both of the aforementioned mutations of concern (MOCs) is required, so as to immediately take the appropriate public health actions.

The guidelines from the European Centre for Disease Prevention and Control (ECDC) and the World Health Organization (WHO) indicate that SARS-CoV-2 complete genome sequencing, or at least, S gene sequencing (whole or partial), should be used to confirm infection with a specific variant (European Centre for Disease Prevention and Control (ECDC), 2021). Undoubtedly, genome analysis via the next-generation sequencing (NGS) technology comprises the most accurate approach for SARS-CoV-2 molecular characterization. However, the application genome sequencing is labor-intensive, costineffective and may take several hours up to days to be completed, depending on the workflow of the laboratory. According to the guidelines of the aforementioned organizations, alternative methods, such as diagnostic screening PCR-based assays can also be used, for the early detection and prevalence calculation of VOCs (European Centre for Disease Prevention and Control (ECDC), 2021). Many of the currently available PCR-based assays are based on the S gene target failure, associated with the allele "drop-out" phenomenon from targeting deletions alone, or in combination with other traits (Kováčová et al., 2021) . As a result, the target of the aforementioned assays is to identify specific VOCs, and not the MOCs associated with the respective amino-acid substitutions possessed by a given variant. On the other hand, SNP-specific assays have been developed, e.g. a melting-curve based assay CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10.1101/2021.05.31.21257367 doi: medRxiv preprint revealed single and double nucleotide mismatches in 25,130 and 206 genomic sequences, respectively, whereas in the case of primer SARSpDo5, only single mismatches were found in 3,327 sequences ( Fig. 1) . Primer SARSpUp2 was evaluated in silico using the DINAMelt software (Markham and Zuker, 2005) , so as to estimate at the annealing temperature of 56 o C, the mole fraction of each oligonucleotide hybridized to the target regions with most common single mismatches found in SARS-CoV-2 genomic sequences ( whereas modification of a guanine nucleotide or either of its nearest-neighbor bases was avoided in G•T mismatch sites. The probes were short (16 bases) to improve mismatch discrimination. Analysis for the melting temperature (Tm), the absence of dimers, and possible hairpin secondary structures was performed using the IDT OligoAnalyzer™ Tool (Integrated DNA Technologies). This tool can also predict stability of LNA-DNA duplexes.

Based on our experience on designing similar assays, all probes targeted the same DNA strand and additional LNAs were incorporated in the probes, if needed, so as their Tm can be calculated to approximately 63.5 °C. The primer responsible for the extension that degrades the hybridized probes was designed to have a Tm 1-2 °C lower, but not less. This procedure for primer design is adopted so as: a) to avoid significant strand extension before the annealing of probes during the PCR cycling and b) to apply annealing temperatures that allow both sufficient mismatch discrimination and sufficient amplification efficiency. The second primer was designed to have a higher Tm in order to support a) high amplification efficiency . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10.1101/2021.05.31.21257367 doi: medRxiv preprint and b) annealing to targets with single mismatches (Table 1, Fig. 1 ) so as to permit high analytical and diagnostic sensitivity of the assay.

The protocol and composition was developed and optimized using EnzyQuest's One step RT qPCR kit (Product No.: RN010; EnzyQuest P.C., Heraklion, Greece). Reactions (20 μ l) were performed using 4 mM Mg 2+ , the aforementioned oligonucleotides at the concentrations presented in Table 1 Subsequently, two ten-fold dilution series were prepared in a background of a SARS-CoV-2negative RNA extract from human oropharyngeal swabs, i.e. from 5×10 5 down to 5×10 1 copies/assay for the WT, and from 10 6 down to 10 2 copies/assay for the South African variant.

All prepared dilutions of both dilution series were run in triplicates to determine the amplification efficiency and the linearity of the developed assay against each targeted variant.

An amplification efficiency of >94% (i.e. FAM: 99.5%; HEX: 98.8%; Texas Red: 94.2%; Cy5: 95.0%) and a linear range of quantification over 5 log 10 were observed in all cases (Fig.   2 ). In order to determine the limit of detection (LOD) both quantified samples were further diluted (between 150 and 12.5 copies/reaction) and tested. Each prepared dilution was tested . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10.1101/2021.05.31.21257367 doi: medRxiv preprint in octaplicate, and the LOD was calculated with 95% probability of detection, by applying probit regression analysis. Consequently, the LOD for the typing assay was determined at 117 copies/reaction. The specificity of the developed assay in testing human samples was assessed by testing a panel of SARS-CoV-2-negative human nasopharyngeal swabs (N=20) originating from a university hospital setting. These swabs (as those comprising the positive panel A CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10.1101/2021.05.31.21257367 doi: medRxiv preprint investigation (B.1.1.318) as only the 484K phenotype was present. Testing of these RNA extracts using the developed assay was performed as described above, on a Rotor-Gene Q 5plex Platform (Qiagen). The channels used for fluorescence acquisition were Green, Yellow, Orange and Red, for HEX, FAM, Texas Red and Cy5 fluorophores, respectively. The Rotor-Gene Q Application Software was used for data analysis, through the function "Allelic Discrimination". Overall, the developed assay was able to accurately identify both MOCs and discriminate between the different tested variants (Table 2) . Low levels of non-specific fluorescence occurred in the Texas Red channel, without affecting the analysis process and the obtained results.

Additionally, given that some animal species are susceptible to SARS-CoV-2, a veterinary collection of samples was tested via the developed method. Specifically, a panel of negative samples originating from cats and minks (N = 3 for each species) was tested, revealing the absence of fluorescence in any of the four channels. SARS-CoV-2-positive oropharyngeal swabs from cats which were investigated in a previous work of our team (Chaintoutis et al., 2021) were also tested. SARS-CoV-2-positive oropharyngeal samples from minks were also tested. Specifically, 3 animals infected by a WT strain (B.1.1.305) originating from a heavily infected farm were tested. All animal samples were correctly identified regarding the detection of the relevant mutations (Table 2 ).

In conclusion, the typing real-time RT-PCR assay developed herein is able to accurately identify the mutations associated with the E484K and N501Y phenotypes of SARS-CoV-2 S protein, considered as MOCs. The assay is simple to perform, rapid, sensitive and can be applied in a variety of specimens from human and animals. Our methodology of primer and probe design can support both high sensitivity of detection and sufficient mismatch discrimination and can facilitate the rapid design of similar assays able to detect emerging MOCs. The results suggest that the developed system is useful as a fast prescreening tool for the selection of VOCs and other variants of interest (VOIs) for subsequent characterization via NGS analysis, or for MOCs prevalence calculation within the framework of epidemiological investigations.

We thank Dr. Anastasia Chatzidimitriou for providing testing samples.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. ; . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 1, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 1, 2021. ; https://doi.org/10. 1101 

Surveillance of SARS-CoV-2 lineage B.1.1.7 in Slovakia using a novel, multiplexed RT-qPCR assay

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

DINAMelt web server for nucleic acid melting prediction

Insights from SARS-CoV-2 sequences

Deep Mutational Scanning of SARS-CoV-2 Receptor Binding Domain Reveals Constraints on Folding and ACE2 Binding

SARS-CoV-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma

Design of LNA probes that improve mismatch discrimination