A high-resolution melt curve toolkit to identify lineage-defining SARS-CoV-2 mutations

Viral RNA samples

Initial development and validation of the assays was performed using four in vitro cultured viruses maintained at LSHTM, for which stocks were obtained from the National Institute for Biological Standards and Controls (NIBSC): 19A SARS-CoV-2 (BetaCoV/Australia/VIC01/2020, CVA-GLA-1, CVA-GLA-2, CVA-GLA-3); two in vitro cultured viruses obtained from BEI resources: Beta VOC (hCoV-19/South Africa/KRISP-EC-K005321/2020), and Gamma VOC (hCoV-19/Japan/TY7-503/2021), and three cultured viruses maintained at LSTM: WT England B1 lineage (hCoV-19/England/20,092,096,704/2020), Alpha VOC (SARS-CoV-2/human/GBR/FASTER_372/2021) and Delta VOC (SARS-CoV-2/human/GBR/Liv_273/2021). In addition, two synthetic SARS-CoV-2 RNA controls (Twist Bioscience, USA) were used for early assay development, which were representative of Alpha VOC (England/205,041,766/2020) and Beta VOC (South Africa/KRISP-EC-K005299/2020).

Clinical samples

A field evaluation to determine assay performance was undertaken using clinical samples collected from participants as part of the Facilitating Accelerated Clinical Validation Of Novel Diagnostics for COVID-19 (FALCON) study at the Liverpool John Lennon airport testing site. Subjects who required COVID-19 RT-PCR testing and who presented to a drive-through regional testing centre in the community were prospectively invited to participate. Informed consent was obtained from all subjects and/or their legal guardian(s). If subjects provided verbal consent, specimens were collected using combined oral and nasopharyngeal swabs (COPAN UTM-RT Diagnostics, Italy) and stored in 1 mL of universal transport media (UTM). Ethical approval was obtained from the National Research Ethics Service and the Health Research Authority (IRAS ID:28,422, clinical trial ID: NCT04408170). All methods were performed in accordance with the relevant guidelines and regulations. Patient specimens were collected using either combined throat and nasal swabs or nasopharyngeal swabs (COPAN UTM-RT Diagnostics, Italy) and stored in 1 mL of universal transport media (UTM).

Samples collected in the FALCON study were confirmed SARS-CoV-2 RNA-positive using the TaqPath™ COVID-19 CE-IVD RT-PCR Kit (ThermoFisher Scientific, USA). Based on epidemiological data were defined as either (i) presumed Alpha (N = 30) if collected between 15th and 22nd January 2021 when Alpha was the most commonly circulating variant in the UK, or (ii) presumed Delta (N = 30) if collected between 14th and 25th June 2021 when Delta was the most commonly circulating variant in the UK. SARS-CoV-2 RNA-negative samples from the FALCON study were used for specificity testing (N = 19). A further 16 samples collected between 13th and 21st December 2021 when Omicron was the most commonly circulating variant in the UK were used in evaluation of the Omicron assay.

RNA extraction

RNA was extracted from in vitro cultured virus and clinical specimens in UTM transport media using the QIAamp Viral RNA Kit (QIAGEN, Germany), following the manufacturer’s protocol, and implemented as either a manual workflow for the viral cultures, or as an automated workflow using the QIAcube HT platform (QIAGEN) for clinical samples. Purified RNA was eluted in 50 μl of elution buffer stored at -20 °C until use.

cDNA synthesis

Extracted RNA was diluted 1:1000 in molecular grade water (Thermo Fisher Scientific, USA) before cDNA generation by random-primed reverse transcription using the SuperScript IV First Strand Synthesis System (Invitrogen, USA). The thermal profile was modified from the manufacturer’s instructions as per the ARTIC PCR protocol; the reverse transcription was carried out at 42 °C for 50 min, followed by 70 °C at 10 min to inactivate the RT enzyme.

Sequencing

Sequencing of SARS-CoV-2 genomes was performed using the ARTIC SARS-CoV-2 sequencing protocol20 on the Oxford Nanopore Technology (UK) MinION device. The ARTIC V3 primer sets were purchased as pools (Integrated DNA Technologies, USA), and the PCR and library preparation were carried out according to the ARTIC V3 sequencing protocol20. For suspected Omicron samples (based on collection date), the updated ARTIC V4.1 primer sets were utilised (Integrated DNA Technologies, USA). All PCR assays used the Q5® Hot Start High-Fidelity 2X Master Mix (New England Biolabs, USA), 10 μM primer pools, and a thermal cycling profile of a 30 s 98 °C heat inactivation, followed by 25 cycles of a 15 s denaturation at 98 °C and a five-minute annealing/extension at 65 °C. Library preparation was carried out using the Ligation Sequencing Kit (SQK-LSK109) and Native Barcoding Expansion Kits (EXP-NBD104 and EXP-NBD114; all Oxford Nanopore Technologies, UK). Enzymes for barcode and adapter ligation were purchased from New England Biolabs (USA), and AMPure XP beads (Fisher Scientific, USA) were used during the library preparation. Sequencing was carried out using an R.9.4.1 flow cell on a MinION device.

Bioinformatics analyses

Basecalling was done via MiniKnow (v4.2.8), with demultiplexing and read filtering using Guppy (v5.0.7.). The ARTIC pipeline20 was then used to assemble a consensus genome, BAM files, and variant calling file with –normalise 200 –threads 4. The pipeline performs a reference alignment of basecalled reads using minimap and aligns the consensus sequence against the reference using Muscle. Automated rapid variant calling was carried out using EPI2ME Desktop Agent v3.3.0 with the ARTIC + NextStrain analysis pipeline, and via the vcf file generated via ARTIC. Genomes were analysed in Tablet (v1.21.02.08)21 for manual inspection of genomes and coverage. All sequence data is deposited in BioProject PRJNA936677 [www.ncbi.nlm.nih.gov/bioproject/PRJNA936677].

HRM primer design

A total of 799 complete SARS-CoV-2 genomes from VOC lineages Alpha, Beta, Delta, Gamma and Kappa, and genomes sequenced from samples collected in England prior to the emergence of the Alpha VOC, were downloaded from GISAID (gisaid.org/). These were aligned using MAFFT version 7.453 with the –auto and –nuc parameters enabled22 alongside the GISAID reference sequence (hCoV-19/Wuhan/WIV04/2019) and all other sequences derived from RNA samples used in initial development. Primer pairs were designed to target seven different SARS-CoV-2 VOC or VUI lineage-defining mutations (Table 1):

  • Orf1ab gene, deletion positions 3675–3677 [Orf1ab_del.3675–3677]

  • Spike gene, deletion positions 156–157 [S_del.156–157]

  • Spike gene, deletion positions 242–244 [S_del.242–244]

  • Spike gene, substitution position 417 [S_K417N]

  • Spike gene, substitution position 484 [S_E484K]

  • Spike gene, substitution position 681 [S_P681H/R]

  • Nucleoprotein gene, substitution position 3 [N_D3L]

Table 1 Description of the SARS-CoV-2 genome targets for, and detail of the sequences of, the seven primer sets designed in this study to amplify variant-defining mutations, and that were subsequently incorporated into multiplex assays A-D. Orf1ab = Open reading frame 1ab, S = Spike, N = Nucleocapsid.

Primer sets were designed either to target conserved sites flanking mutations of interest, or to directly bind the mutation with the 3’ base of the primer. In case of the former, a shift in melt temperature (Tm) is detected if the mutation is present, whereas in the latter, the presence of the mutation prevents primer binding and ablated amplification. Primers were designed using Primer3Plus (www.bioinformatics.nl/cgi-bin/primer3plus/primer3plus.cgi), OligoCalc (biotools.nubic.northwestern.edu/OligoCalc.html) nearest neighbour method was used to estimate amplicons’ Tm23, all primer pairs were analysed for specificity using Primer-BLAST24. Final primer sequences are detailed in Table 1.

HRM assays

Initially, three multiplex HRM assays, each containing three different primer pairs were developed and evaluated in this study. These assays identified combinations of variant-defining mutations for SARS-CoV-2 VOCs, with a focus on differentiation of Alpha from 19A, and later Delta from Alpha, to understand the epidemiology of these variants at the time in the UK.

Each assay was performed using 2.5 μL of RNA template and in 12.5 μl final reaction volumes, using the SuperScript™ III One-Step RT-PCR System with Platinum™ Taq DNA Polymerase kit (ThermoFisher, USA), with final reagent quantities as follows: 1X reaction mix, 0.25 μL of SuperScript™ III RT/Platinum™ Taq Mix, 1X EvaGreen® dye (Biotium, USA), and primers added to their optimised concentration (Table 2).

Table 2 Optimised final reaction primer concentrations for primer set in each multiplex assay, A-D. Primer set names as described in Table 1.

Reactions were performed using the RGQ 6000 5-plex HRM platform (Qiagen, Germany) with the following thermal cycle profile: reverse transcription at 50 °C for 15 min, initial denaturation at 95 °C for 5 min, followed by 40 cycles of 95 °C for 10 s, 56 °C for 30 s and 72 °C for 20 s. HRM was then performed, melting from 73 °C to 85 °C, acquiring data to the HRM channel in 0.1 °C increments, with a 2 s stabilisation between each step. For each assay a reference control of 19A strain SARS-CoV-2 RNA, confirmed by WGS to be negative for all mutations of interest, and a no template negative control were included.

Analysis of HRM assay data

Primary data were analysed using the RGQ system software (v2.3.5, Build 1, QIAGEN) and additional data analyses were performed in Prism (v9, GraphPad, USA).

Data was visualised as negative first derivative plots (Fig. 1), and Tm values were recorded for each peak. Thresholds for analysis were determined empirically during initial optimisation experiments to be as follows:

  • Orf1ab_del.3675–36,770.40 dF/dT

  • S_del.156–1570.35 dF/dT

  • S_del.242–2440.20 dF/dT

  • S_K417N0.20 dF/dT

  • S_E484K 0.43 dF/dT

  • S_P681H/R0.20 dF/dT

  • N_D3L0.20 dF/dT

Figure 1
figure 1

High resolution melt profiles (derivative, dF/dT) of multiplex assays A–D: (A) Multiplex A, targeting S_del.156–157, S_K417N and N_D3L, (B) Multiplex B, targeting S_E484K, S_del.242–244 and S_P681H/R, (C) Multiplex C, targeting Orf1ab_del.3675–3677, S_del.242–244 and S_P681H/R, and (D) Multiplex D, targeting S_del.156–157, S_K417N, N_D3L and S_EPE (all mutations and primer sets as described in Table 1). Data shows evaluation in vitro cultured SARS-CoV-2 strains, where the variant type had been determined by WGS, as follows: 19A (green line), Alpha VOC (purple line) SARS-CoV-2/human/GBR/FASTER_372/2021, Beta VOC (red line) hCoV-19/South Africa/KRISP-EC-K005321/2020, Gamma VOC (black line) hCoV-19/Japan/TY7-503/2021, Delta VOC (blue line) SARS-CoV-2/human/GBR/Liv_273/2021 and Omicron VOC RNA derived from a clinical specimen where SARS-CoV-2 RNA was detected at Ct < 30 and with known complete genome sequence.

For primer sets targeting the S_del.156–157 and S_E484K, ORF1ab_del.3675–3677, primer-dimer curves present below 75 °C were excluded.

Assay interpretation details are shown in Table 3. The Tm threshold for calling a result based on a Tm shift was set by calculating the average Tm for all samples confirmed negative for the mutation and then adding two standard deviations in the direction of the temperature shift (Fig. 2).

Table 3 Validated melt temperature (Tm) values (°C) for peak Tm used in interpretation of high-resolution melt data that indicate sequences positive or negative for the targeted mutation, and threshold boundaries for determining positivity, by primer set (as described in Table 1). * In Omicron VOC, del3674-3676.
Figure 2
figure 2

Melt temperature (Tm, °C) value distributions (with 95% confidence intervals) for clinical samples collected 15th–22nd January 2021 (Alpha, N = 30) or collected between 14 th and 25th June 2021 (Delta, N = 30) where SARS-CoV-2 RNA was detected at Ct < 30 and with known complete genome sequences. Data are shown for each primer set (as described in Table 1): (A) S_del.156–157, (B) S_del.242–244, (C) S_K417N, (D) S_E484K, (E) S_P681H/R (F) Orf1ab_del.3675–3677, and (G) N_D3L. Blue dotted lines represent the positivity thresholds (shown in Table 3), which when crossed indicates the presence of the mutation. Positivity thresholds were calculated by the addition or subtraction of two standard deviations surrounding the mean Tm, of samples confirmed negative for the mutation.

Rapid development of primers for the emerging Omicron BA.1 variant

In response to the emergence of the Omicron BA.1 VOC during the HRM assay development, a primer set was designed targeting the lineage-defining insertion of three amino-acid residues, glutamate-proline-glutamate (EPE) at position 214 in the spike protein in Omicron BA.1 lineage14. Primers were designed so the forward primer sits on the insertion and amplifies when the mutation is present. The Omicron primers were first optimised as a singleplex assay and evaluated using 16 samples confirmed as Omicron BA.1 by sequencing, alongside ten Alpha and ten Delta SARS-CoV-2 samples. The primers were designed to be compatible with Multiplex A, and their compatibility was demonstrated using single RNA samples of each VOC.The target of the Omicron primer set, ins214EPE, is only present in the Omicron BA.1 lineage, and further primer sets would be required in order to detect other Omicron lineages. This is potentially facilitated by the presence of unique mutations in all Omicron sublineages so far encountered25. This is highlighted to illustrate the speed at which this assay can be updated to accommodate emerging VOCs.

Read more here: Source link