Optimization and Clinical Validation of Colorimetric Reverse Transcription Loop-Mediated Isothermal Amplification, a Fast, Highly Sensitive and Specific COVID-19 Molecular Diagnostic Tool That Is Robust to Detect SARS-CoV-2 Variants of Concern

Abstract

The coronavirus disease 2019 (COVID-19) pandemic unfolded due to the widespread severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission reinforced the urgent need for affordable molecular diagnostic alternative methods for massive testing screening. We present the clinical validation of a pH-dependent colorimetric reverse transcription loop-mediated isothermal amplification (RT-LAMP) for SARS-CoV-2 detection. The method revealed a limit of detection of 19.3 ± 2.7 viral genomic copies/μL when using RNA extracted samples obtained from nasopharyngeal swabs collected in guanidine-containing viral transport medium. Typical RT-LAMP reactions were performed at 65°C for 30 min. When compared to reverse transcriptase-quantitative polymerase chain reaction (RT-qPCR), up to cycle-threshold (Ct) value 32, RT-LAMP presented 98% [95% confidence interval (CI) = 95.3-99.5%] sensitivity and 100% (95% CI = 94.5-100%) specificity for SARS-CoV-2 RNA detection targeting E and N genes. No cross-reactivity was detected when testing other non-SARS-CoV virus, confirming high specificity. The test is compatible with primary RNA extraction-free samples. We also demonstrated that colorimetric RT-LAMP can detect SARS-CoV-2 variants of concern and variants of interest, such as variants occurring in Brazil named gamma (P.1), zeta (P.2), delta (B.1.617.2), B.1.1.374, and B.1.1.371. The method meets point-of-care requirements and can be deployed in the field for high-throughput COVID-19 testing campaigns, especially in countries where COVID-19 testing efforts are far from ideal to tackle the pandemics. Although RT-qPCR is considered the gold standard for SARS-CoV-2 RNA detection, it requires expensive equipment, infrastructure, and highly trained personnel. In contrast, RT-LAMP emerges as an affordable, inexpensive, and simple alternative for SARS-CoV-2 molecular detection that can be applied to massive COVID-19 testing campaigns and save lives.

Keywords: COVID-19; RT-LAMP; SARS-CoV-2; diagnostic test; molecular test; respiratory virus.

FIGURE 1

Reverse transcription loop-mediated isothermal amplification (RT-LAMP) for SARS-CoV-2 RNA detection and COVID-19 testing. Inactivated saliva samples or from nasopharyngeal swabs can processed for RNA extraction previously or be directly added to RT-LAMP reaction. Colorimetric output can be achieved by different sensors and can be read by naked eye. The whole procedure is rapid and simple and does not require complex infrastructures. Created with biorender.com.

FIGURE 2

Analytical sensitivity as revealed by the limit of detection (LoD). RNA was extracted from VTM-nasopharyngeal swab, and the genome viral copies input was calculated based on SARS-CoV-2 E gene-harboring plasmid (Bioclin #K228-1) calibration curve. RT-LAMP reaction was performed at 65°C during 30 min using WarmStart^® colorimetric master LAMP mix (NEB #M1800) in 20 μL final volume (upper panel). Amplicons were resolved in 2% agarose gel and stained with GelRed^® (Biotium #41003) to confirm DNA amplification (bottom panel). cps/μL, viral genome copies per microliter; NTC, nontemplate control; VTM, viral transport medium (Bioclin #G092-1).

FIGURE 3

Colorimetric RT-LAMP for COVID-19 diagnosis validation using 100 clinical samples. Clinical samples were collected from symptomatic and hospitalized patients by nasopharyngeal swabs in partnership with CT-Vacinas/UFMG, Belo Horizonte, Brazil. Samples were obtained from different parts including Brazilian Southeast and Northeast regions. The reaction was performed at 65°C during 30 min using WarmStart^® colorimetric LAMP master mix (NEB #M1800) in 20 μL final volume. The RT-LAMP reaction targeted SARS-CoV-2 N gene. Yellow content indicates positive reaction, whereas the pink pattern reveals nonreagent samples. Amplicons were resolved in 2% agarose gel and stained with GelRed^® (Biotium #41003) to confirm DNA amplification. Latter pattern confirmed specific SARS-CoV-2 amplification that matches with yellow output tubes, which is not observed in pink nonreagent tests. +C, positive control using RNA extracted from laboratory-Vero E6 cultured inactivated SARS-CoV-2; NTC, nontemplate control. Clinimetric parameters from these samples are presented in Supplementary Figure S1.

FIGURE 4

Colorimetric RT-LAMP for SARS-CoV-2 RNA detection. (A) Box-and-whisker representation of colorimetric RT-LAMP SARS-CoV-2–positive and –negative output (x axis) plotted in function of their respective RT-PCR Ct values (y axis). Forty-eight false negative samples were detected on RT-LAMP after Ct 32 despite other 55 being positive from Cts ranging from 32 to 39. (B) Receiver operating characteristic (ROC) curve constructed based on data presented in A. As summarized in Table 1, high-sensitivity values were obtained at the predicted cutoff.

FIGURE 5

Microbial cross-reactivity assay to test SARS-CoV-2 RT-LAMP analytical sensitivity. The test was performed using potentially cross-reacting respiratory viruses or local occurring arboviruses. RT-LAMP reaction was performed at 65°C during 30 min, with additional 10 min, to confirm the absence of cross-reactivity when targeting SARS-CoV-2 E and N genes. The assay was performed using the WarmStart^® colorimetric LAMP 2× master mix (NEB #M1800). Yellow (positive) reaction is observed only when the template is SARS-CoV-2 viral RNA. hRSV, human respiratory syncytial virus; NTC, nontemplate control; M, molecular size marker. RT-LAMP amplification products were resolved in 2% agarose gel and stained with GelRed^® (Biotium #41003) to confirm DNA amplification. DENV3, dengue virus serotype 3; ZIKV, Zika virus; CHIKV, Chikungunya virus; YFV, yellow fever virus; Influenza A (H1N1/H3N2); and influenza B (Yamagata/Victoria).

FIGURE 6

Colorimetric RT-LAMP for SARS-CoV-2 detection using genes N, E, and RdRp as target. Selected SARS-CoV-2–positive clinical samples by RT-qPCR were classified as low (Ct 18.9 and 21.7), medium (Ct 26.6 and 28.4), and high (Ct 31.6 and 35.2) Ct values for E gene. They were included as input for colorimetric RT-LAMP reaction using primers targeting N, RdRp (A), and E genes (B). RT-LAMP SARS-CoV-2 false-negative samples were more frequent when using E and RdRp genes as target (C). RT-LAMP reaction was performed at 65°C during 30 min, using the WarmStart^® colorimetric LAMP 2× master mix (NEB #M1800). RT-LAMP amplification products were resolved in 2% agarose gel and stained with GelRed^® (Biotium #41003) to confirm DNA amplification. +C, positive control using SARS-CoV-2 RNA extracted from laboratory-cultured inactivated SARS-CoV-2; NTC, nontemplate control.

FIGURE 7

Colorimetric RT-LAMP to detect SAR-CoV-2 in RNA extraction–free clinical samples (A) or laboratory-cultured virus (B). Clinical samples were derived from nasopharyngeal swabs placed on guanidine-containing viral transport medium, diluted 1:10. The RT-PCR Ct values for SARS-CoV-2 based on E gene are as follows: CS134 = 31.8, CS135 = 15.3, CS138 = 18.4, CS139 = 21.7, and CS140 = 24.6. RT-LAMP reaction was performed in 20 μL final volume, incubated at 65°C during 30, 40, or 50 min (inactivated virus) using WarmStart^® colorimetric LAMP master mix (NEB #M1800). Both clinical samples and viruses are RNA extraction–free samples. The amplification products (amplicons) were migrated in agarose gel at 2% to confirm amplification, as indicated by the characteristic ladder highlighted by GelRed^® staining. NTC, nontemplate control; CS, clinical sample; and +C, positive control.

FIGURE 8

Colorimetric RT-LAMP allows the detection of SARS-CoV-2 VOCs and VOIs. RT-LAMP reaction was performed at 65°C for 30 min, using the WarmStart^® colorimetric LAMP 2× master mix (NEB #M1804), using multiplex N2/E1 primer sets. The amplicons were migrated in agarose gel at 2% to confirm amplification, as indicated by the characteristic ladder highlighted by GelRed^® staining. NTC, nontemplate control; CS, clinical sample; and +C, positive control. The top panel shows a schematic representation of SARS-CoV-2 spike protein (upper) and where the main mutations are highlighted and represented in SARS-CoV-2 virions (right hand side) present in VOC gamma (B.1), delta (B.1.167.2), and VOI zeta (P.2). The VOCs alpha (B.1.1.7) and beta (B.1.3.51), first reported in the United Kingdom and South Africa, respectively, are also represented. K417N: lysine-to-asparagine substitution at position 417 of spike protein at the receptor biding domain (RBD); V445A: valine-to-alanine substitution at position 445 and so on. L, leucine; Q, glutamine; E, glutamic acid; Y, tyrosine; T, threonine; P, proline; H, histidine; D, aspartic acid; S, serine; F, phenylalanine. del, deletion. Segments of SARS-CoV-2 protein NTD, N-terminal domain; CTD2, C-terminal domain 2 or C terminus of S1 fragment after furin cleavage; FP, fusion peptide; HR1, heptad repeat region 1. SARS-CoV-2 variants were previously sequenced. Variants of interest B.1.1.371 and B.1.1.374 were first reported in Saudi Arabia and Finland, respectively, (https://cov-lineages.org/). Created with biorender.com.

FIGURE 9

Schematic representation of SARS-CoV-2 genome indicating the amplicons for the COVID-19 molecular diagnostics by RT-LAMP. Structural representation of SARS-CoV-2 virion shows the main particle parts. LAMP primer regions are indicated as previously reported (Baek et al., 2020; Ben-Assa et al., 2020; Butler et al., 2020; Chow et al., 2020; Dudley et al., 2020; Song et al., 2021; Ganguli et al., 2020; Huang et al., 2020; Lamb et al., 2020; Lu et al., 2020; Mohon et al., 2020; Park et al., 2020; Rabe and Cepko, 2020; Thi et al., 2020; Yan et al., 2020; Yu et al., 2020; Zhang et al., 2020a,b; Anahtar et al., 2021; Bhadra et al., 2021; Bokelmann et al., 2021; González-González et al., 2021). ORF, open reading frame; RdRp, RNA-dependent RNA polymerase; NSP, nonstructural protein. Schematic representation created using Snap Gene Viewer software version 5.0.7; N1, N2, and N3_CDC correspond to the amplicons for SARS-CoV-2 detection by RT-PCR. Created with biorender.com.