Starting from scratch: Step-by-step development of diagnostic tests for SARS-CoV-2 detection by RT-LAMP

Abstract

The pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has affected millions of people worldwide. Public health strategies to reduce viral transmission are based on widespread diagnostic testing to detect and isolate contagious patients. Several reverse transcription (RT)-PCR tests, along with other SARS-CoV-2 diagnostic assays, are available to attempt to cover the global demand. Loop-mediated isothermal amplification (LAMP) based methods have been established as rapid, accurate, point of care diagnostic tests for viral infections; hence, they represent an excellent alternative for SARS-CoV-2 detection. The aim of this study was to develop and describe molecular detection systems for SARS-CoV-2 based on RT-LAMP. Recombinant DNA polymerase from Bacillus stearothermophilus and thermostable engineered reverse transcriptase from Moloney Murine Leukemia Virus were expressed using a prokaryotic system and purified by fast protein liquid chromatography. These enzymes were used to set up fluorometric real time and colorimetric end-point RT-LAMP assays. Several reaction conditions were optimized such as reaction temperature, Tris-HCl concentration, and pH of the diagnostic tests. The key enzymes for RT-LAMP were purified and their enzymatic activity was determined. Standardized reaction conditions for both RT-LAMP assays were 65°C and a Tris-HCl-free buffer at pH 8.8. Colorimetric end-point RT-LAMP assay was successfully used for viral detection from clinical saliva samples with 100% sensitivity and 100% specificity compared to the results obtained by RT-qPCR based diagnostic protocols with Ct values until 30. The developed RT-LAMP diagnostic tests based on purified recombinant enzymes allowed a sensitive and specific detection of the nucleocapsid gene of SARS-CoV-2.

Fig 1. Expression and purification of recombinant Bacillus stearothermophilus DNA polymerase (Bst) and reverse transcriptase (RT) from Moloney Murine Leukemia Virus.

Coomassie blue-stained 8% Tricine-SDS-PAGE electrophoresis gel analysis. (A) Expression of recombinant Bst enzyme and Ni²⁺-IMAC purification. Lane 1: clarified supernatant loaded; Lane 2: flow-through fraction; Lane 3–6: washing step fractions; Lane 7–9: eluted fractions containing Bst. (B) Heparin column purification of recombinant Bst. Lane 1: desalted sample loaded; Lane 2: flow-through fraction; Lane 3: washing step fraction; Lane 4–6: elution fractions; Lane 7–8: concentrated Bst-containing fractions. (C) Final Bst formulations from three different purification batches (B1, B2, B3). (D) Expression of recombinant RT enzyme and Ni²⁺-IMAC purification. Lane 1–3: flow-through fractions; Lane 4–5: washing step fractions; Lane 6–10: elution fractions containing RT. (E) Cation exchange column purification of recombinant RT. Lane 1: IMAC elution fraction; Lane 2: desalted sample loaded; Lane 3–5: flow-through fractions; Lane 6–7: washing step fractions; Lane 8–13: elution fractions. (F) Final RT formulations from three different purification batches (B1, B2, B3). M: molecular weight marker; C+: previously purified Bst or RT enzyme, employed as control positive; I: insoluble fraction; S: soluble fraction; C: clarified supernatant. Violet or green arrows indicate the expected size for Bst and RT enzymes, respectively.

Fig 2. Optimization assays of RT-LAMP reaction conditions for SARS-CoV-2 detection.

(A) Real time LAMP reactions evaluating the effect of temperature on amplification. (B) Real time RT-LAMP reactions at different pH and concentration of Tris-HCl. (C) Colorimetric RT-LAMP reactions at pH 8.8 in the absence of Tris-HCl in the reaction buffer. The figure shows the electrophoretic profile of the amplification reaction products (upper panel) and the colorimetric determination of each reaction (lower panel). NTC: non-template control; M: DNA molecular weight maker 1 Kb Plus DNA Ladder (Invitrogen). Shadows represent standard deviation from triplicate curves. Different lowercase letters (a, b) indicate significant differences among treatments based on Dunn’s test at p < 0.05.

Fig 3. Analytical sensitivity and specificity of RT-LAMP.

(A) Real time RT-LAMP amplification plots obtained from ten-fold serial dilutions of in vitro N1 transcript ranging between 10⁹ and 10² copies/reaction. (B) Colorimetric change and electrophoretic amplifications obtained from ten-fold serial dilutions of in vitro N1 transcript ranging from 10¹¹ to 10² copies/reaction. (C) Evaluation of specificity through the optimized fluorometric real-time RT-LAMP assay with the recombinant Bst and RT enzymes against respiratory controls NATROL-1 and NATROL-2. NTC: non-template control. M: DNA molecular weight marker 1 Kb Plus DNA Ladder (Invitrogen). N1 transcript: 1×10⁹ copies of in vitro transcribed N1 gene. Reference material: 2 μL of AccuPlex SARS-CoV-2 reference material.

Fig 4. Evaluation of colorimetric end-point RT-LAMP assays for SARS-CoV-2 detection in clinical samples.

RNA was extracted from saliva samples and evaluated for SARS-CoV-2 detection. Comparison of Ct values from RT-qPCR protocols and colorimetric end-point RT-LAMP (Colorimetric RT-LAMP) results from 100 samples tested. Negative samples (Ct > 40 and undetermined) for SARS-CoV-2 detection by RT-qPCR are shown at Ct = 40. (A) 59 samples were called positive by CDC RT-qPCR protocol-based methodology using 2019-nCoV-N1 primers/probe set (y-axis) and were compared to colorimetric RT-LAMP results (x-axis) detected as positive (yellow) or negative (pink). (B) 34 samples were called positive by Berlin RT-qPCR protocol-based methodology using RdRp_SARSr-P2 probe (y-axis) and were compared to colorimetric RT-LAMP results (x-axis) detected as positive (yellow) or negative (pink). Dashed lines at the top of each panel serve as reference for samples below the detection limit for RT-qPCR.