RT-LAMP primers
Several published SARS-CoV-2 RT-LAMP primer sets14,15,36,37,38 were carefully screened in terms of the detection sensitivity, false positive and false negative rates, reaction speed, and test reproducibility (Figs. S1–S6). The best performing primer set38 (Table 2) was selected for further characterization and optimization in our lyophilized colorimetric RT-LAMP home test kit. This primer set targets the ORF1a gene of the SARS-CoV-2 viral genome and is minimally impacted by mutations on recent SARS-CoV-2 variants of concern. An additional primer set23 (Table S2) was tested to confirm the reliable performance of the one-pot lyophilization protocol.
RT-LAMP reagents
WarmStart Colorimetric LAMP 2X Master Mix (New England Biolabs, cat. M1800L) was used as the RT-LAMP master mix for the test kit. RT-LAMP primers were ordered from IDT as custom DNA oligos with standard desalting. The primers were resuspended in nuclease-free water (Sigma-Aldrich) and mixed to form a 10X primer mix consisting of 2 µM F3 primer, 2 µM B3 primer, 16 µM FIP primer, 16 µM BIP primer, 4 µM LoopF primer, and 4 µM LoopB primer. RT-LAMP reactions were run at 20 µL total reaction volume. Specifically, the lyophilized RT-LAMP reagents were reconstituted with 5 µL sample + 15 µL nuclease-free water in all analytical experiments conducted with synthetic SARS-CoV-2 RNA control (Twist Bioscience, cat. 102024). Unless otherwise specified, 20 µL of sample (as opposed to 5 µL sample + 15 µL nuclease-free water) was directly added to the lyophilized RT-LAMP mix in validation experiments conducted with simulated SARS-CoV-2 infected samples.
Fast one-pot lyophilization of colorimetric RT-LAMP
The 3 M trehalose solution was prepared by dissolving 0.5 g D-(+)-trehalose dihydrate powder (Sigma-Aldrich, M.W. 378.33 g/mol) in 440.5 μL nuclease-free water, followed by vigorous vortexing and heating at 60 °C for 10 min to fully dissolve the trehalose to yield a supersaturated solution. This resulted in a solution with a total volume of approximately 760 μL, corresponding to an effective trehalose concentration of around 1.75 M. Because our assay prototyping needed only small amounts of the solutions, to keep the measuring simple and consistent, we refer to this resulting solution as the 3 M trehalose throughout the text unless otherwise specified. The trehalose solution was then sterilized by filtering through a 0.2 µm syringe filter (VWR), followed by brief vortex and centrifuge to remove air bubbles. For consistency, the 1 M GuHCl solution was prepared similarly by directly dissolving 0.1 g guanidine hydrochloride powder (VWR, M.W. 95.53 g/mol) in 1046.8 μL nuclease-free water without further adjustment of the final volume. The tube containing the GuHCl solution was covered with aluminum foil to protect it from light. Components of the colorimetric RT-LAMP lyophilization formulation were mixed at the specified ratio (Table 1), aliquoted into 0.2 mL PCR tubes, and frozen at − 20 °C for 1 h. Finally, the tubes were quickly transferred with caps open into a vacuum concentrator (Savant Speedvac SVC-100H) connected to the lyophilizer (VirTis Freezemobile 12SL). Lyophilization was run for 1 h with the chamber pressure at ~ 10 milliTorr and the condenser temperature at ~ − 40 °C. Details of the RT-LAMP lyophilization protocol optimizations are shown in Fig. S7–S16.
Viruses
SARS-CoV-2 Pango lineage A (WA-1/2020) was used in all whole virus assays for in-house test kit optimization. Other variants obtained through BEI Resources were B.1.1.7, B.1.351, B.1.526.2 and B.1.617.1. Other viruses were isolated from clinical samples, and lineage determined by next generation sequencing. These include P.2, B.1.2, B.1.427, B.1.429, B.1.526, and B.1.617.2. Further virus information including source are summarized in Table S3. Authentic SARS-CoV-2 were propagated and inactivated in a BSL-3 laboratory under a protocol approved by the OHSU Institutional Biosafety Committee under the supervision of Dr. Tafesse.
Simulated SARS-CoV-2 infected samples
Anterior nasal swab (and alternatively gingival swab) samples were collected as described below from 10 uninfected individuals, into sample collection media (TBE (1X: 90 mM Tris–borate-2 mM EDTA, pH 8.0) at the specified concentration, in nuclease-free water) and pooled. Heat-inactivated SARS-CoV-2 virus was diluted in sample collection media to achieve the specified viral copy numbers, according to in-house qRT-PCR. Virus-spiked samples were heated at 95 °C for the specified duration and chilled on ice. Then, 20 µL of the resulting sample was added directly into the RT-LAMP microtube to reconstitute the lyophilized reagents. RT-LAMP microtubes were vortexed (spun down, if necessary) and briefly chilled on ice, before pre-incubation photos were taken. Finally, RT-LAMP microtubes were incubated for 60 min at 65 °C, with photos taken at 30–60 min to assess color change. The best detection sensitivity was achieved by RNA isolation with 0.01X TBE (based on 5 µL sample input into a 20 µL reaction) and heating for 10 min at 95 °C (Figs. S17–18). Subsequent assays were performed using 20 µL direct sample input and 0.0025X TBE, to reduce pipetting steps.
Colorimetric RT-LAMP in thermocycler
RT-LAMP microtubes containing samples or non-template control (NTC) were vortexed, spun down and briefly chilled on ice before pre-incubation photos were taken. RT-LAMP microtubes were incubated in a thermocycler for 60 min at the specified temperature, with photos taken at 30–60 min to assess color change. Tubes were briefly chilled on ice to allow color stabilization, before being photographed. The optimal RT-LAMP incubation temperature was identified by running the reactions with a temperature gradient (T = 65 °C, G = 5 °C) set in a gradient thermal cycler (Eppendorf MasterCycler). To avoid contamination, the RT-LAMP tubes should never be reopened after the incubation reaction.
Colorimetric RT-LAMP in thermos
Both the viral RNA preparation and the RT-LAMP incubation were conducted in a thermos. Freshly boiled water was added to pre-warm the thermos for 2 min and then dumped out. Next, boiling water was re-added into the thermos and chilled to ~ 97 °C, after which the virus-spiked samples and NTC (swab media without virus) were incubated for 10 min in the thermos (with lid on) and then chilled on ice for 5 min to allow cell debris to settle. Next, 20 µL of the heat-inactivated sample “supernatant” (i.e., no cell debris) or NTC were transferred to the RT-LAMP microtube using a disposable transfer pipette. The microtubes were recapped and flicked gently (being careful not to introduce bubbles) to resuspend the lyophilized RT-LAMP reagents and then chilled on ice before pre-incubation photos were taken. Meanwhile, a mug was filled with boiling water, and allowed to chill to ~ 70 °C before pouring into the thermos. The water was allowed to further chill to ~ 67 °C before samples were added. Next, the RT-LAMP microtubes were incubated in the thermos with the lid tightly closed for 60 min, with photos taken at 30–60 min. During incubation, the microtubes were secured on a foam floater to ensure that they were vertically and sufficiently submerged in water to activate the RT-LAMP reaction. Finally, the tubes were removed from thermos and briefly chilled on ice to allow color stabilization, before being photographed for test result readout.
Quantitative RT-PCR (OHSU)
Heat-inactivated SARS-CoV-2 RNA was isolated using the Zymo Directzol RNA purification kit according to the manufacturer’s protocol and eluted in 50 μL elution buffer. SARS-CoV2 RNA levels were measured by a one-step quantitative real time reverse transcription polymerase chain reaction assay (qRT-PCR) using TaqMan One-Step RT-PCR Master Mix (Applied Biosystems) with 4 μL per reaction. Primers and probes were as follows: Forward: 5’-TTTGGCTTTGTGTGCTGACTCT; Reverse: 5’-CCCTTTGAGTGCGTGACAAAT and TaqMan probe: 5’ FAM-ATTGGTGGAGCTAAAC-MGB. Forward and reverse primers were used at 250 nM in the reaction, and the probe at 200 nM. For RNA standards, a ten-fold dilution series of 10^6 to 10^1 of a synthetic RNA control (Twist Biosciences: MN908947.3) was used.
Statistical analysis
To estimate the analytic sensitivity of the RT-LAMP platform with human samples, simulated NP swab samples were spiked with serial dilutions of inactivated WA-1 strain of SARS CoV-2 starting at a calculated 17,473 copies, twofold to 8737 copies and then every 1.6-fold to a final dilution of 227 viral copies. Three independent replicates four-reactions per dilution were performed for each dilution. The average percent-positive for each dilution was calculated and fitted to a dose–response curve with log RNA copy as the dose and percent positive samples as the response and subjected to Find ECanything least squares dose–response curve fitting with lower and upper limits of 0 and 100 respectively (GraphPad Prism V9.3). Fifty and ninety-five percent sensitivity estimates were estimated by setting the F parameter to 50 and 95 respectively.
Thermoses
Three different types of thermoses were used in these assays, available from Amazon (ASINs B00IR77HMW (#1), B08LPZZGCT (#2), B07MJR3P1H (#3)). Temperature drift experiments were conducted—thermoses were pre-warmed and filled with 67 °C water, lids were secured, and final temperatures taken after 40 min, on 3 separate days. Final temperatures were 61.5 °C, 62.5 °C and 62.5 °C for thermos #1, 61 °C, 65 °C and 65 °C for thermos #2, and 61 °C, 62 °C and 62 °C for thermos #3.
Instruction for sample self-collection using anterior nasal swab
No food or drink other than water 30 min prior to sample collection. Wash hands prior to sample collection. Insert swab into nostril just enough so the cotton tip is no longer visible. Swipe the inside of nostril in a circular motion, 5 times. Repeat for the other nostril, using the same swab. Dunk the swab into the labeled tube, plunging it into the liquid about 10 times. Discard the swab and replace the lid on the tube.
Instruction for sample self-collection using gingival swab
No food or drink other than water 30 min prior to sample collection. Wash hands prior to sample collection. Insert bristled swab into mouth and position the swab so that it covers the gingival line (line between gum and teeth) of top teeth. Gently swipe back and forth several times along the gingival line (in a tooth-brushing motion) on the outside face of top teeth. Flip the swab over and repeat for the outside face of bottom teeth. Dunk the gingival swab into the labeled tube, plunging it into the liquid about 10 times. “Squeeze” out the remaining fluid in the swab by pressing it on the side of the tube like a sponge. Discard the swab and replace the lid on the tube.
Human research ethics
This study has been reviewed and approved by the Oregon Health and Science University (OHSU) Institutional Review Board (IRB#20114). Informed consent was obtained from subjects upon enrollment.
