key: cord-1028811-1smrkien
authors: Garneret, Pierre; Coz, Etienne; Martin, Elian; Manuguerra, Jean-Claude; Brient-Litzler, Elodie; Enouf, Vincent; González Obando, Daniel Felipe; Olivo-Marin, Jean-Christophe; Monti, Fabrice; van der Werf, Sylvie; Vanhomwegen, Jessica; Tabeling, Patrick
title: Performing point-of-care molecular testing for SARS-CoV-2 with RNA extraction and isothermal amplification
date: 2021-01-11
journal: PLoS One
DOI: 10.1371/journal.pone.0243712
sha: 8e0827c9b7eaff11826d9943c9be5bce30ede84b
doc_id: 1028811
cord_uid: 1smrkien

To respond to the urgent need for COVID-19 testing, countries perform nucleic acid amplification tests (NAAT) for the detection of SARS-CoV-2 in centralized laboratories. Real-time RT—PCR (Reverse transcription—Polymerase Chain Reaction), used to amplify and detect the viral RNA., is considered, as the current gold standard for diagnostics. It is an efficient process, but the complex engineering required for automated RNA extraction and temperature cycling makes it incompatible for use in point of care settings [1]. In the present work, by harnessing progress made in the past two decades in isothermal amplification and paper microfluidics, we created a portable test, in which SARS-CoV-2 RNA is extracted, amplified isothermally by RT—LAMP (Loop-mediated Isothermal Amplification), and detected using intercalating dyes or fluorescent probes. Depending on the viral load in the tested samples, the detection takes between twenty minutes and one hour. Using a set of 16 pools of naso-pharyngal swab eluates, we estimated a limit of detection comparable to real-time RT-PCR (i.e. 1 genome copies per microliter of clinical sample) and no cross‐reaction with eight major respiratory viruses currently circulating in Europe. We designed and fabricated an easy-to-use portable device called “COVIDISC” to carry out the test at the point of care. The low cost of the materials along with the absence of complex equipment will expedite the widespread dissemination of this device. What is proposed here is a new efficient tool to help managing the pandemics.

Nucleic Acid Amplification Tests (NAATs) detect the presence of pathogen genomes in infected samples through specific amplification of theirnucleic acids,. They are characterized by a low limit of detection (LoD),allowing detection of viral loads as low as 1-100 genome copies per microliter of sample, and an excellent analytical specificity. There are different types of NAATs for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing but today,

PLOS ONE | https://doi.org/10.1371/journal.pone.0243712 January 11, 2021 1 / 9 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

RT-PCR (Reverse Transcription Polymerase Chain Reaction), coupled to extraction, is considered as the gold standard for diagnostic laboratories. For getting high-throuput, RT-PCR, due to an inherent complexity has to be performed in parallelized tests in centralized laboratories. This generates logistics issues, long delivery times, (several days in practice) and excessive costs for developing countries. Over the last ten years, with the advent of isothermal amplification technologies [2] , hope has been raised to perform NAATs at the point of care, with portable devices much cheaper than extraction/RT-PCR platforms, and with comparable performances.

In this context, NAATs coupling isothermal amplification to paper microfluidics have been developed in laboratories [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] . These systems possess a potential, but owing a number of limitations (moderate practicability, absence of performing sample preparation in most cases, incomplete evaluation of the clinical and analytical performances) these devices have not reached a state, yet, where they can be commercialized. Thereby, as SARS-CoV-2 started to propagate, the technology was not ready to face the request of large scale global testing. Recently, commercial isothermal detection assays have been proposed [17] . However, because of substantial cost, limited throughput and moderate performances, they have not taken off. Isothermal detection of SARS-CoV-2has been reported in the literature, but these assays do not represent yet an alternative to PCR testing, often because they do not include the extraction of the nucleic acids (NA), which affects their sensitivity [18] [19] [20] [21] .

The molecular tests we report here are portable and low cost. The new feature is that they combine NA extraction, RT-LAMP and naked eye visualization capability on the same device. Teir performances are comparable to the extraction/RT-PCR method. The practical device we created (COVIDISC) is adapted for the detection of SARS-CoV-2 at the point of care.

The device is shown on Fig 1A. Two sheets of black polypropylene (Polydis, 800μm thick), covered on one side by PCR tape (thermal Seal RT2R, Sigma Aldrich) two centimeters in size, 4 -Folding of the device in such a way that the extraction membrane comes into contact with the two reaction disks. On both disk is freeze-dried the LAMP mix and primers (Covid-19 and human 18S RNA for the test and control disk respectively) permiting reverse transcription and amplification. 5 -Elution of the RNA from the capture membrane to the reaction disks, then, sealing with PCR tape 6 -Heating at 65˚C. Read-out in real time with an intercalating agent (SYTO82).

https://doi.org/10.1371/journal.pone.0243712.g001 incorporate pretreated laser-cut membranes: one (oblong form: 2mm x 10mm) for nucleic acid extraction, and the two others (circular form: 5.4 mm diameter) for the RT-LAMP reactions. Capture membrane and reaction membranes are binder-free glass fiber provided by Cytiva. COVIDISC devices are 3D-printed using a Poly lactic acid (PLA) resin, printed at 115˚C at 90mm/second speed, 0.15 mm layer thickness. The diameter of the COVIDISC is 65 mm.

RT-LAMP primer sequences have been retrieved from published work [22] [23] [24] [25] . The freeze dried reaction mix is composed of 8U GspSSD 2.0 polymerase (OptiGene), 0.5 U AMV-RT (Life Sciences Advanced Technologies), 4 mM MgSO4, 0.4 mM of each dNTPs (dATP, dTTP, dGTP, dCTP), 5μM SYTO-82 intercalating dye (Invitrogen), 0.2 μM of F3/B3 primers, 0.8 μM FIP/BIP primers, 0.4 μM of LF/LB primers, 1M Betain and 10% trehalose. Volumes of 18μL reaction mix are pipetted onto reaction membranes and freeze as previously described in [24] .

AVL buffer was used for sample lysis (QIAGEN). Rinsing buffer was made according to previously described recipe [23] . Elution was performed using isothermal reaction buffer (iBuffer) provided by OptiGene. For sample and washing steps, a cellulose absorbant pad (Whatman CF7) was used for downstream capillary pumping.

Naso-pharyngal swabs collected from suspected COVID-19 cases and eluted in Universal Viral Transport medium (UTM) were provided by Institut Pasteur biobank. To prepare a reference set of SARS-CoV-2 positive samples, swab eluates presenting similar SARS-CoV-2 viral loads (genome quantities) were mixed together to obtain eight pools with concentrations spanning the range of viral loads detected in COVID-19 patient samples, as measured with the real-time duplex qRT-PCR (targeting the RdRP gene) developed by the French National Reference Centre [26, 27] . In terms of viral loads, the pools span a range of RT-PCR cycle threshold (Ct) values extending from 15 to 38, i.e from 10 8 to 1 copies per microliter of clinical sample (See S3 Fig in S1 File for the conversion between Ct values and viral load.). Each pool was tested in duplicate to determine the limit of detection of the method. To determine the reaction specificity, we used eight additional pools of clinical naso-pharyngal samples, not infected by SARS-CoV-2, but collected from patients infected by other respiratory pathogens: respiratory syncytial virus (RSV) A and B, influenza A (H1) virus, influenza A (H3) virus, influenza B Victoria (BVIC), Yamagata lineages (BYAM), human rhinovirus (HRV) and human metapneumovirus (hMPV). Swab eluates presenting high viral loads (above 1000 copies/μL) were used to prepare these pools. In addition, 5 SARS-CoV-2 negative naso-pharyngal eluates were used as control. Fig 1B shows the workflow. The sample is mixed with a lysis buffer. The role of this buffer is both to lyze the virus and enable the nucleic acid extraction process. The principle of the nucleic acid extraction was described by Boom [28] . Essentially, the sample is mixed with a salt and ethanol buffer. Concentrated salt inverts the z-potential of the matrix (from negative to positive). With that inversion, the negatively charged nucleic acids (RNA and DNA) are captured. The other sample components (proteins, membranes, etc.) are precipitated by ethanol and transported downstream, through the extraction disk. In practice, 100 μL of the sample, are mixed with 200μL AVL buffer (Qiagen), and incubated for 10 minutes. A volume of 200μL pure ethanol is then added and the solution is introduced in the extraction membrane and washed with a washing buffer according to following reference [23, 29, 30] . In order to manage the buffers flowing through the capture membrane, a wick is placed in contact with it to pump the fluids by capillarity. The absorbent is further thrown away. After drying for 15 minutes, the device is folded so as to place the extraction membrane in contact with the two reaction membranes. RNA is eluted by using Optigene standard isothermal buffer (iBuffer). In this process, the buffer releases the captured RNA from the extraction membrane, and, owing to the action of capillarity, penetrates in the two reaction disks. On both disks, the freeze-dried reaction mixis hydrated, and mixes with the RNA extracts. Upon temperature rise to 65˚C (by maintaining the paper device on a hot plate), reverse transcription and amplification are performed. We use real-time RT-LAMP. Its kinetics is measured by tracking the fluorescence emission of a DNA intercalating dye (SYTO-82). The target is Orf1ab (primers are displayed in S1 Table in S1 File [19] ) and ARN 18S as internal sample processing control and sample quality control. Measuring the kinetics of the amplification allows to estimate the noise level and control the shapes of the amplification curves. The noise level should be low (a deficient extraction often generates noise on such curves) and the amplification curves should be sigmoidal. Any deviation from these expectations may signal the presence of an artifact. Indeed, for on-field applications, kinetic studies are not necessary. Only end-point data (i.e after one hour amplification) is needed for providing a YES/NO answer. A specific request for an information derogation has been submitted to a competent national ethics committee for the tested control samples for which provision of such information for the secondary use of the samples could not be ascertained. ambigously in the lower one. We conclude that the limit of detection of the method is around 1 copy per microliter of initial sample (i.e. detected at a Ct value of 38 by RT-PCR). Such an analytical sensitivity is comparable to real-time RT-PCR performances. These conclusions do not sensibly depend on the image treatment. For instance, the value of 25% we took above, for determining the intensity levels on Fig 2, is not critical. We varied this number between 10 and 50%, without significant change in the results. Sup Mat 4 shows a classification based on the fluorescence standard deviation, which leads to the same results. Naked eye could also be used to distinguish positive from negative samples. The method of treatment of the images is thus not critical.

The insert of Fig 2 shows the kinetics of five RT-LAMP reactions, for different viral load levels, using SYTO82 as fluorescent intercalating agent. The RT-LAMP reaction kinetics obtained on a sample from a healthy individual has been represented, for estimating the background noise [31] . This noise level corresponds to the natural fluorescence of the reaction disk, i.e approximately 800 arbitrary intensity units. The fluorescence intensities of the SARS--CoV-2-positive samples are located well above. They have the expected sigmoid shape. The sample with the highest viral load, (presenting a Ct value of19, equivalent to 8x10 6 genome copies per microliter of sample), was detected after 15 minutes. At higher Ct values, i.e. as viral load decreases, the signal takes off at later times. Plateaus are observed for the patient sample pools 1-6, i.e up to a Ct value of 33. For the samples with the lowest viral loads (pools 7 and 8), i.e., 31 and 1 genome copies per microliter, a plateau is not fully reached, but still, the virus can be detected without ambiguity.

The right part of Fig 2 addresses the question of the specificity. The images show that for non-SARS-CoV-2 pathogens, in almost all cases, fluorescence is barely visible. Sample H1 is seen above the background, for unclear reasons. However, quantitatively, the difference with the average background is 20%, which is not significant . Fig 2 shows that the specificity of the test, based on this set of pathogens, is 100%.

In order to transform the foldable paper system into a practical POC device, we created the "COVIDISC", shown in Fig 3 . Fig 3A shows the COVIDISC workflow, which exactly reproduces that of Fig 1B ( see caption of Fig 3A) . The structure of the device, shown in Fig 3B consists of two plastic disks, 5 cm in diameter, able to rotate around a common axis. The extraction membrane, the wick and the S2 Table in S1 File) . On adding QUASR probes (Quenching of Unincorporated Amplification Signal Reporters) [32] to RT-LAMP, we obtain a naked eye YES/NO answer. Fig 3B (right) shows the readout of RNA extracts of SARS-CoV-2, captured with the camera of a smart phone. We used an oven to maintain the temperature at 65˚C and one LED and two gelatin filters to take the picture (see S4 Fig in S1 File) . The equipment needed for running the test is thus minimal.

To the best of our knowledge, the system we present here is unique. No such assay, i.e. portable, low-cost, including NA extraction/elution, user friendly and with performances comparable to RT-PCR, has been reported in the literature thus far. The results presented here have been obtained with 21 clinical samples and larger sets of samples are certainly desirable. Nonetheless, our results indicate that, with a minimal equipment, one can extract, wash, elute, reverse-transcribe, amplify and measure the kinetics, with a LoD comparable to the gold standard real-time RT-PCR, i.e. 1 genome copies per microliter of clinical sample, and a specificity, of 100%, based on the set of pathogens we have used. We also created a portable device (COVI-DISC) that can be used at the point of care. The production cost we estimated for such a device is around 2-4 $. This type of device is thus appropriate for ressource-limited countries but also for developed countries, to reduce the now considerable costs generated by the generalization of RT-PCR testing in the population. Generalizing these tests at the doctor's practice, at the working place or in pharmacies could allow the isolation of infected patients without delay, shortening their quarantine, reducing logistics and costs and offering a new efficient approach to manage the pandemics. In the future, obviously, the same technology could be used for other pathogens.

Supporting information S1 File. The supplementary information file contains information about the primer sequences (S1), the internal positive controls (S2), the protocol of operation for the COVI-DISC (S3), the signal processing used (S4), the relation between the Ct and the viral load (S5) and the COVIDISC design (S6). (DOCX)

Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. InCold Spring Harbor symposia on quantitative biology

Point-of-care nucleic acid testing for infectious diseases

Determination of nitrite in saliva using microfluidic paper-based analytical devices

Measuring markers of liver function using a micropatterned paper device designed for blood from a fingerstick. Analytical chemistry

Field evaluation of a prototype paper-based point-of-care fingerstick transaminase test. PloS one

A rapid and automated device for purifying nucleic acids

Understanding was printing: a simple micropatterning process for paper-based microfluidics

Paper-based molecular diagnostic for Chlamydia trachomatis

Patterned paper as a platform for inexpensive, low-volume, portable bioassays. Angewandte Chemie

Paper based RNA detection and multiplexed analysis for Ebola virus diagnostics. Scientific reports

Paper microfluidics for nucleic acid amplification testing (NAAT) of infectious diseases

Isothermal amplification using a chemical heating device for point-of-care detection of HIV-1. PloS one

Paper machine" for molecular diagnostics

based microfluidics for DNA diagnostics of malaria in low resource underserved rural communities

A fully disposable and integrated paper-based device for nucleic acid extraction, amplification and detection

A smartphone-based diagnostic platform for rapid detection of Zika, chikungunya, and dengue viruses. Scientific reports

A reverse-transcription loop-mediated isothermal amplification (RT-LAMP) assay for the rapid detection of SARS-CoV-2 within nasopharyngeal and oropharyngeal swabs at Hampshire Hospitals NHS Foundation Trust. medRxiv

Rapid colorimetric detection of COVID-19 coronavirus using a reverse tran-scriptional loop-mediated isothermal amplification (RT-LAMP) diagnostic plat-form: iLACO. medRxiv

Rapid Detection of Novel Coronavirus (COVID19) by Reverse Transcription-Loop-Mediated Isothermal Amplification. Available at SSRN 3539654

High-surety isothermal amplification and detection of SARS-CoV-2, including with crude enzymes. bioRxiv

SARS-CoV-2 Detection Using an Isothermal Amplification Reaction and a Rapid. Inexpensive Protocol for Sample Inactivation and Purification

Loop-mediated isothermal amplification of DNA. Nucleic acids research

Paper-Origami-Based Multiplexed Malaria Diagnostics from Whole Blood

Lyophilized visually readable loop-mediated isothermal reverse transcriptase nucleic acid amplification test for detection Ebola Zaire RNA

LAMP user guide mastermixes and assay optimisation

WHO Coronavirus disease COVID-19 technical guidance: Laboratory testing for 2019-nCoV in humans

Protocol: real-time RT-PCR assays for the detection of SARS-CoV-2, Institut Pasteur

Rapid and simple method for purification of nucleic acids

A simple and rapid DNA extraction method from whole blood for highly sensitive detection and quantitation of HIV-1 proviral DNA by real-time PCR

A low cost point-of-care viscous sample preparation device for molecular diagnosis in the developing world; an example of microfluidic origami

Reverse Transcription-Loop-mediated Isothermal Amplification (RT-LAMP) Assay for Zika Virus and Housekeeping Genes in Urine, Serum, and Mosquito Samples

Quenching of unincorporated amplification signal reporters in reverse-transcription loop-mediated isothermal amplification enabling bright, single-step, closed-tube, and multiplexed detection of RNA viruses. Analytical chemistry