key: cord-0889484-zxlvxz1h
authors: Banada, Padmapriya; Green, Raquel; Banik, Sukalyani; Chopoorian, Abby; Streck, Deanna; Jones, Robert; Chakravorty, Soumitesh; Alland, David
title: A Simple Reverse Transcriptase PCR Melting-Temperature Assay To Rapidly Screen for Widely Circulating SARS-CoV-2 Variants
date: 2021-09-20
journal: J Clin Microbiol
DOI: 10.1128/jcm.00845-21
sha: d94dd0b593ed68eebc49426e47db78e5d4b83d8f
doc_id: 889484
cord_uid: zxlvxz1h

The increased transmission of SARS-CoV-2 variants of concern (VOC), which originated in the United Kingdom (B.1.1.7/alpha), South Africa (B1.351/beta), Brazil (P.1/gamma), the United States (B.1.427/429 or epsilon), and India (B.1.617.2/delta), requires a vigorous public health response, including real-time strain surveillance on a global scale. Although genome sequencing is the gold standard for identifying these VOCs, it is time-consuming and expensive. Here, we describe a simple, rapid, and high-throughput reverse transcriptase PCR (RT-PCR) melting-temperature (T(m)) screening assay that identifies the first three major VOCs. RT-PCR primers and four sloppy molecular beacon (SMB) probes were designed to amplify and detect the SARS-CoV-2 N501Y (A23063T) and E484K (G23012A) mutations and their corresponding wild-type sequences. After RT-PCR, the VOCs were identified by a characteristic T(m) of each SMB. Assay optimization and testing was performed with RNA from SARS-CoV-2 USA WA1/2020 (wild type [WT]), B.1.1.7, and B.1.351 variant strains. The assay was then validated using clinical samples. The limit of detection for both the WT and variants was 4 and 10 genomic copies/reaction for the 501- and 484-codon assays, respectively. The assay was 100% sensitive and 100% specific for identifying the N501Y and E484K mutations in cultured virus and in clinical samples, as confirmed by Sanger sequencing. We have developed an RT-PCR melt screening test for the major VOCs that can be used to rapidly screen large numbers of patient samples, providing an early warning for the emergence of these variants and a simple way to track their spread.

COVID-19 vaccines (2, 13, 14) . All three variants, B.1.1.7, B.1.351, and P.1, contain the N501Y mutation, while the South African and Brazilian variants additionally contain mutations in E484K and K417N (1, 12) . Thus, the N501Y mutation appears to be an excellent marker for all three strains, while the E484K mutation can be used to differentiate the other two strains from B.1.1.7. New SARS-CoV-2 variants have also been recently reported in the United States (5, 15) . Most of the new variants also have mutations at the 501 and/or 484 position, although additional mutations, such as L452R, have also been reported in more recent variants (15, 16) ; however, the epidemiological and clinical relevance of these new variants are not well understood.

All three of the major SARS-CoV-2 N501Y variants are circulating around the world (1). However, the distribution, extent, and spread of these variants are poorly understood in countries such as the United States, where only a small fraction (0.04 to 3.5%) of COVID-19 cases are analyzed by viral genomic sequencing per the CDC's national genomic surveillance dashboard (17) . Viral genome sequencing is currently the only method available to reliably detect rapidly emerging SARS-CoV-2 variants. Genomic sequencing has the advantage of providing a detailed map of new mutations, which supports new variant discovery as well as monitoring the exact type of the variant. However, genomic sequencing is expensive and difficult to perform in real time. In contrast, reverse transcriptase PCR (RT-PCR) testing for SARS-CoV-2 has become widespread. This diagnostic approach is easy to perform in a high-throughput manner, and rapid turnaround times are possible. However, routine RT-PCR tests do not differentiate among SARS-CoV-2 variants or do so only by producing negative assay results (18) , which still require sequence confirmation (19) . Furthermore, the potential for additional mutations to appear near key variant-defining alleles may complicate the development of RT-PCR assays for SARS-CoV-2 variants (20) . We have previously demonstrated that sloppy molecular beacons (SMBs) combined with melting temperature (T m ) code analysis may be used to specifically detect mutations in short genomic regions where a variety of mutations can exist (21) .

Here, we apply this same T m -based approach to detect and differentiate variant strains of SARS-CoV-2 with high sensitivity and specificity. This approach is flexible and can be used in a high-throughput manner for screening and detection of SARS-CoV-2 variants, easily allowing the addition of new mutation-detecting assays as needed to identify and track new variants as they emerge. Furthermore, this approach can be performed on a wide range of real-time PCR instrumentation as long as they have the capacity to run melting curve analysis. The wide availability of such instruments can allow quick adoption of this assay around the world, increasing access to real-time monitoring of SARS-CoV-2 variant spread. . RNA extracted from all three variant strains was quantified against a standard curve generated with the N1 gene-specific real-time RT-PCR assay (22, 23) , and the concentration was determined to be 4 Â 10 5 GE/ml for B.1.1.7, 7 Â 10 6 GE/ml for B.1.351, and 8.5 Â 10 6 GE/ml for the P.1 variant strains.

Genome sequence analysis for assay design. For initial analysis of the mutations in SARS-CoV-2, a total of 330,132 high-quality viral genome sequences deposited in GISAID (24) as of 12 January 2021, were analyzed. Publicly available data sets were analyzed in this study. These data can be found at https://www.gisaid.org/. A 250-nucleotide region around the N501Y (A23063T) and E484K (G23012A) positions in the reference strain (GenBank accession number MN908947) was selected and used to identify the corresponding regions in the GISAID data set using BLAST (25) . These matching sequences were condensed into a set of unique sequences and aligned using a multiple-sequence alignment program, MAFFT (26) . Candidate amplification primers and probes were identified on the basis of sequence conservation and predicted T m using the algorithm of SantaLucia (27) , and a final set of primers was designed with the help of the Primer3 program (28) . SMB probe design was performed using the web servers DNA mfold (http://www.unafold.org/mfold/applications/dna-folding-form.php) and DINAmelt (http://www.unafold.org/hybrid2.php) to predict the probe folding structures and probe-target hybrid T m values, respectively.

Primers and probes. The list of primers and probes used for the 501 variant (SMB-501) and the 484 variant (SMB-484) assays are shown in Table 1 . Both assays were run as two separate reactions. For the SMB-501 assay, an 89-bp region surrounding the 23063 position was amplified using an asymmetric PCR. For SMB-484 assay, a 76-bp region surrounding the 23012 position was amplified using an asymmetric PCR. SMB were designed targeting both the wild-type 501N/484E (23063A/23012A) and the mutant 501Y/484K (23063T/ 23012G) sequences (SMB-501-MT). Primers were obtained from Millipore Sigma (The Woodlands, TX), and SMBs were synthesized by LGC Biosearch Technologies (Petaluma, CA). An internal control (IC) assay developed by the CDC (22, 23) , targeting the human RNaseP gene, was simultaneously performed for each extracted RNA specimen as a separate reaction in a separate well, using the TaqMan real-time PCR assay probe tagged with 6-carboxyfluorescein (FAM) at the 59 end and Dabcyl quencher at the 39 end.

SMB assay formulation and procedure. TaqPath 1-step RT-quantitative PCR (qPCR) master mix CG (ThermoFisher Scientific, Waltham, MA) was used for the RT-PCR. Each one-step reaction mix was supplemented with 0.2 mM the forward primer 501-F/484-F and 2 mM/4 mM the reverse primer 501-R/484-R, 0.4 mM each of the SMB probes (SMB-501-WT and SMB-501-MT/SMB-484-WT and SMB-484-MT), and 1 ml of the template RNA (note that the concentration of the forward primers did not match the concentration of the reverse primers so as to create an asymmetric PCR). The internal control contained primers and a probe specific for human RNaseP as described previously (22, 23) . Each reaction was run in replicates of 4 in 384-well plates in a Roche LightCycler 480 (Roche, Indianapolis, IN). The one-step RT-PCR amplification was performed with the following thermocycling conditions: uracil DNA glucosylase incubation for 2 min at 37°C and reverse transcription (RT) for 15 min at 50°C, followed by asymmetric PCR for 45 cycles (denaturation at 95°C for 1 s, annealing/extension at 55°C for 30 s). The post-PCR melt was performed with the following conditions: denaturation at 95°C for 30 s, followed by cooling down to 45°C and gradual heating to 85°C, with continuous monitoring of fluorescence at a rate of 2 acquisitions per degree Celsius. The total assay time was 77 min. Automated T m calls were performed by the LC480 T m detection software at the end of the PCR. The resulting T m for each probe was identified and matched with the T m signature code defined for the wild type or the mutant variants.

Analytical sensitivity. The prequantitated genomic RNA from the SARS-CoV-2 USA WA1/2020 (WT), B.1.1.7 (alpha/UK variant), and B.1.351 (beta/SA variant), obtained from BEI Resources, were diluted in Tris-EDTA (TE) buffer. Total RNA was extracted from a SARS-CoV-2 negative nasopharyngeal (NP) specimen (confirmed negative by Xpert Xpress SARS-CoV-2 test). The diluted SARS-CoV-2 RNA from the WT and the variants was spiked into the extracted matrix at final concentrations of 1,000, 500, 100, 50, 10, 5, and 1 GE/ml and mixed well. A 1-ml aliquot of the RNA spiked into the NP matrix was added to 19 ml of the one-step RT-PCR mix containing the primers and probes and was evaluated in the SMB-501 assay or the SMB-484 assay; 1,000 GE/ml of the quantified RNA from P.1 was also validated with both assays ( Table 2) .

Validation with patient samples. Deidentified NP swabs obtained from patients undergoing routine COVID-19 clinical testing using the Xpert Xpress SARS-CoV-2 or Xpert Xpress SARS-CoV-2/Flu/RSV (4-plex) test (Cepheid, Sunnyvale CA) assay in the CLIA-and CAP-certified laboratory at the Public Health 

Limit of detection. The one-step RT mix containing asymmetric PCR primers and both wild-type and MT-specific probes was added with either WT RNA or MT RNA at different concentrations, ranging from 1,000 to 1 GE/reaction for both SMB-501 and SMB-484 assays. After T m analysis was performed, the T m peak heights were highest in the assays with the largest added number of GE, and the peak heights progressively decreased as the number of GEs present in the reactions decreased. However, T m peak heights produced by both SMB 501-WT/SMB 484-WT and SMB 501-MT/SMB 484-MT could still be reproducibly detected at concentrations down to 10 GE/reaction when tested against both WT and N501Y mutant strains, defining the assay limit of detection as #10 GE per reaction for both the SMB-501 and SMB-484 assays ( Fig. 1 ; see also . Any samples that failed to produce a T m value for either of the SMBs or produced T m values outside the range defined for the T m codes would have been defined as indeterminate and the assay repeated. Using this code definition, we retested our reference RNA samples 20 times, and the correct 501/484 allele was detected in each case, achieving an analytic sensitivity and specificity of 100%. These results clearly indicate that the combination of both SMB 501-WT and SMB 501-MT probes can specifically detect and differentiate the N501Y variants from the wild-type strains with high confidence.

Validation with patient samples. A total of 76 patient samples, 46 confirmed COVID-19 positive and 30 confirmed COVID-19 negative, were tested using the SMB-501 assay. A subset of 45 of these same samples (26 COVID-19 positive and 19 COVID negative) were additionally tested with the SMB-484 assay. None of the COVID-19-negative samples produced any measurable T m values from either of the assays, yielding a specificity of 100%. Positive samples were selected based on a wide range of N2-C T values between 16 and 41 based on the Xpert Xpress SARS-CoV-2 assay. As shown in Table 3 , 40/45 (89%) of the COVID-19-positive samples produced measurable T m values for both SMB probes in the SMB-501 assay in at least one of the two replicates. Two samples were identified as indeterminate based on the inability of either the WT or MT probe to generate a measurable T m and/or due to the presence of a failed internal control. Five clinical samples that had previously tested positive for SARS-CoV-2 generated negative results in our variant assay, in that neither probe produced a measurable T m . All of the samples with negative assay results had previously shown initial Xpert C T values of $35. Although many of the samples with Xpert C T values as late as 39 could still be detected by our assay, we found that the samples with a C T of $35 using the Xpert assay yielded relatively stunted melting peak heights of #0.3 with WT-SMB and #0.1 with MT-SMB, similar to that observed when testing ,10 GE in our limit-of-detection studies, which is indicative of very low viral loads in these samples. However, the T m could still be identified for at least one of the 501 SMB probes in all samples, except the negatives. Overall (Table 3 , Fig. 2 ). Twelve representative WT and mutant samples, as predicted by the SMB assay, underwent Sanger sequencing of the PCR products for confirmation of the PCR results. In all cases, the WT or mutant sequences identified by the SMB N501Y assay were confirmed by the sequencing (Fig. 3) .

The emergence of SARS-CoV-2 VOC with the potential for increased transmission, disease severity, and resistance to vaccine-induced immunity is of grave concern (14, 29). A simple screening assay to monitor the emergence and spread of these strains may be helpful for implementing public health strategies to counter these and future strains. Our study demonstrates that our assay is simple, rapid, and sensitive and specific for detecting key variant-identifying mutations using a high-throughput PCR assay platform. Thus, this assay has the potential for relatively inexpensive high-throughput testing for rapid identification of N501Y, E484K variants. In designing this assay, we took advantage of the fact that the N501Y and E484K mutations are common in the major SARS-CoV-2 variants. Since this mutation also appears to be responsible for the increased infectivity and possibly the other adverse manifestations of these strains (4), assays that detect this mutation may also prove useful to detect any future strain that evolves to have increased transmission potential.

Our assay is the first, to our knowledge, that uses post-PCR T m -based analysis to detect and differentiate SARS-CoV-2 variants using SMB probes demonstrated in clinical samples. This assay format has the benefit of producing a measurable T m result irrespective of whether the SMB probe is fully complementary to its target nucleic acid sequence. Instead of detecting a mutation by either producing or not producing a signal, SMBs detect mutations by producing a T m shift. Failure to produce a T m signal indicates an invalid assay rather than the absence of a mutation. The robustness of our assay is further increased by our use of two different SMBs, one complementary to the WT sequence and one complementary to the mutant sequence. The pattern of T m values or T m signature produced by the combined T m values of each SMB probe provides an unequivocal identification of a WT or MT sequence. We have also shown that T m signatures can be used to detect mixtures of mutant and WT sequences and to identify numerous mutations present in an assay's target region (29) . Thus, we expect that our assay should continue to be able to identify N501Y and E484K variants even if additional mutations develop near this primary mutation within the probe-footprint region once the specific T m signatures of each new genotype are characterized. Our assay is meant to be a screening assay that will identify samples likely to contain SARS-CoV-2 variants of concern. Although in this study we have proposed the use of a combination of two probes per assay for increased accuracy, we expect that most SARS CoV-2 variants can be identified using only the MT probe alone, and future versions of our assay are likely to use a single SMB to identify mutations that are less critical than N501Y and E484K. We suggest that a useful public health strategy would be to screen COVID-19-positive samples in near real time with our assay and then to perform genomic sequencing on a subset of screen-positive samples. This sequencing would confirm the presence of the expected variants and, in some cases, lead to the discovery of new variants. Ongoing sequencing of a subset of all SARS-CoV-2 samples will also be required to identify completely novel variants or to investigate the epidemiology and clinical characteristics of variants such as those recently reported in the United States. Fortunately, our assay is easily extensible, and additional tests for new key mutations can be added in a modular format to a screening panel when new mutations associated with critical new variants are discovered. In fact, it is our intention to continuously update this assay until our work is superseded by a better approach or much more widespread genome sequencing becomes commonplace. At the time of this assay development and manuscript preparation, additional strains originated in United States (epsilon) and India (delta) were not considered VOCs. We are currently optimizing and screening samples for these additional mutations. Key updates will also be posted on a preprint journal. In the meantime, we hope that our current test will help increase surveillance and potentially help control the spread of the new emerging variants of concern.

Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.6 MB. SUPPLEMENTAL FILE 2, XLSX file, 11.6 MB.

Emergence of SARS-CoV-2 B.1.1.7 lineage-United States

Genetic variants of SARS-CoV-2-what do they mean?

Covid-19: new UK variant may be linked to increased death rate, early data indicate

The high infectivity of SARS-CoV-2 B.1.1.7 is associated with increased interaction force between Spike-ACE2 caused by the viral N501Y mutation

Novavax offers first evidence that COVID vaccines protect people against variants

CDC. 2021. Emerging SARS-CoV-2 variants

2021. mRNA-1273 vaccine induces neutralizing antibodies against spike mutants from global SARS-CoV-2 variants

Comprehensive mapping of mutations in the SARS-CoV-2 receptor-binding domain that affect recognition by polyclonal human plasma antibodies

Public health actions to control new SARS-CoV-2 variants

Early analysis of a potential link between viral load and the N501Y mutation in the SARS-COV-2 spike protein

Early transmissibility assessment of the N501Y mutant strains of SARS-CoV-2 in the United Kingdom

Genomic evidence of SARS-CoV-2 reinfection involving E484K spike mutation

Escape from neutralizing antibodies by SARS-CoV-2 spike protein variants

COVID-19 vaccines vs variants-determining how much immunity is enough

Emergence of a novel SARS-CoV-2 variant in southern California

Reduced sensitivity of infectious SARS-CoV-2 variant B.1.617.2 to monoclonal antibodies and sera from convalescent and vaccinated individuals

National genomic surveillance dashboard

Multiplexed RT-qPCR to screen for SARS-COV-2 B.1.1.7 variants: preliminary results

RNA sequencing: advances, challenges and opportunities

Classification and specific primer design for accurate detection of SARS-CoV-2 using deep learning

Comparative evaluation of sloppy molecular beacon and dual-labeled probe melting temperature assays to identify mutations in Mycobacterium tuberculosis resulting in rifampin, fluoroquinolone and aminoglycoside resistance

Real-time reverse transcription-polymerase chain reaction assay for SARS-associated coronavirus

Saliva or nasopharyngeal swab specimens for detection of SARS-CoV-2

Data, disease and diplomacy: GISAID's innovative contribution to global health

Gapped BLAST and PSI-BLAST: a new generation of protein database search programs

MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform

A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics

Primer3-new capabilities and interfaces

Molecular dynamic simulation reveals E484K mutation enhances spike RBD-ACE2 affinity and the combination of E484K, K417N and N501Y mutations (501Y.V2 variant) induces conformational change greater than N501Y mutant alone, potentially resulting in an escape mutant

This study was funded by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under award number R01 AI131617. D.A. receives research support and royalty payments from Cepheid, which sells the Xpert Xpress SARS-CoV-2 and Xpert Xpress SARS-CoV-2/Flu/RSV tests. S.C. is an employee of Cepheid, and R.J. is a paid consultant for Cepheid.The following reagent was deposited by the Centers for Disease Control and Prevention and obtained through BEI Resources, NIAID, NIH: genomic RNA from SARSrelated coronavirus 2, isolate USA-WA1/2020, NR-52285. The following reagent was obtained through BEI Resources, NIAID, NIH: severe acute respiratory syndrome-related coronavirus 2, isolate hCoV-19/England/204820464/20200, NR-54000, contributed by Bassam Hallis.We thank all the laboratories that contributed sequence data to the GISAID EpiCoV database. A GISAID acknowledgment table reporting the geographic origin and contributions of genomes analyzed in this study is included in the supplemental material as Table S1 .