Rapid, multiplex detection of SARS-CoV-2 using isothermal amplification coupled with CRISPR-Cas12a

Abstract

In December 2019 an outbreak erupted due to the beta coronavirus Severe Acute Respiratory Syndrome Coronavirus 2 in Wuhan, China. The disease caused by this virus (COVID-19) rapidly spread to all parts of the globe leading to a global pandemic. Efforts to combat the pandemic rely on RT-qPCR diagnostic tests that have high turnaround times (~ 24 h), are easily contaminated, need specialized equipment, facilities, and personnel that end up increasing the overall costs of this method. Loop-mediated isothermal amplification (LAMP) coupled with a reverse transcription step (RT-LAMP) is an alternative diagnostic method that can easily overcome these obstacles, when coupled with CRISPR/Cas it can eliminate false positives. Here we report a fast (~ 40 min), highly sensitive, point-of-care multiplex RT-LAMP and CRISPR/Cas12a assay to detect SARS-CoV-2. This fluorescence-based test achieved 100% specificity and 93% sensitivity using 25 positives and 50 negative patient samples for Ct < 35. Our reported LoD of 3 copies/µL will enable the robust, fast detection of the virus in a dedicated equipment which is a major step towards population-wide accessible testing.

Figure 1

Real-time fluorescence for the NEB RT-LAMP reactions using the N/E primer set (a), the As1 primer set (c) and the N/E/As1 combined primer set (e). End point fluorescence measurements for the respective primer sets (b), (d) and (f). Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.

Figure 2

(a) Real-time fluorescence signal for the amplification of synthetic RNA controls using a 3-primer set with 40 mM of Guanidine HCL in the RT-LAMP reaction. (b) Amplification of synthetic RNA controls using a 3-primer set with 40 mM of Guanidine HCL and extra Bst 2.0 polymerase.

Figure 3

Detection of Sars-CoV-2 in the Cas12a reaction (2 step reaction). (a). End point fluorescence of the non-optimized reaction and the respective kinetics (b). (c) Detection of individual genes amplified by RT-LAMP by Cas12a using the optimized version and respective kinetic profile (d). (e) Limit of detection using a 3-primer set with guanidine hydrochloride and with extra Bst 2.0 polymerase. Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.

Figure 4

Melting curve analysis of the RT-LAMP reaction. Individual genes were amplified via RT-LAMP using different amounts of synthetic RNA resulting in the corresponding melting curve using E primers (a), N primers (b), As1 primers (c) and multiplex amplification (d). This analysis was used to validate the successful amplification in the RT-LAMP reaction using an intercalating dye from 35 to 95 °C with a heating ramp of 0.5 °C /s.

Figure 5

Limit of detection using the 2 step RT-LAMP/Cas12a reaction as recommended by the FDA. (a) End point detection of the different synthetic viral loads. (b) 20 replicates at this concentration revealed the detection of 19/20 indicating that this is the number of copies that is detected 95% of the time and thus our limit of detection. Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.

Figure 6

Melting curve analysis from the RT-LAMP reaction of clinical samples. Samples that were considered SARS-CoV-2 positive by INSA were further analysed. Melting curve for sample ADR35 (a), ADR31 (b) and ADR14 (c) using the primers sets for the 3 different genes. (d) End-point fluorescent signal measured in the Cas12a reaction. Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.

Figure 7

Inactivation of nucleases in saliva samples and limit of detection of spiked samples. RnaseAlert substrate nuclease detection system reaction for the different saliva samples using the SHINE buffer (a) and the Tween-20 buffer (b) in a 1:1 proportion. (c) End-point fluorescence signal in the Cas12a-mixture of spiked saliva samples processed using the SHINE buffer. Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.

Figure 8

One-pot RT-LAMP-CRISPR/Cas12a detection of synthetic RNA. End point RFU signal using a dedicated equipment (a) after 10 min and Varioskan Lux multimode plate reader. (b) Statistical significance was determined by unpaired two-tailed t-test and all data were shown as mean ± S.D of 3 replicates. Asterisks indicate **P < 0.01; ***P < 0.001; ****P < 0.0001 and “ns” is non-significant.