• 
  
• # proj-neb-lamp-test, # proj-nucleic-acid-amplification, # chlg-detection-diagnostic
• Google folder with supplementary materials for the project: https://drive.google.com/open?id=1ahzaeP2Z8Kh3TW0qkF1ijUBgGAvR7yZX. See also https://www.facebook.com/onehourcovidtest.
• See Documents section of this project for more supplementary materials including protocols, literature and budgets.  
• Please note that there have been three project updates: one in May 2020; one in October 2020; and one in March 2021. Details are given near the end of this document.

Introduction 

Project title One Hour Covid Test (formerly Optimization of the NEB LAMP Test)

Short name 1hr Covid Test

Note: Reviewers for Round 5 Grant Proposals, please see Project application - project 163 - 21_03_2021.pdf in the Documents tab.  This proposal contains the latest updates for the project. A short stand-alone progress update summary for March 2021 can also be found in the Documents tab.

Problem and Background (200 words max)

There is a shortage SARS-CoV-2 testing throughout the world. Test protocols that diagnose infections or detect environmental contamination by the virus are not yet simple, affordable or widely accessible. Currently most diagnostic testing is accomplished by Reverse Transcription–quantitative Polymerase Chain Reaction (RT-qPCR). RT-qPCR is expensive, laborious and not scalable or accessible to many people living in low-resource settings. Over-reliance on this complicated technique of molecular amplification limits people in many regions of the globe from knowing whether the virus is present and hinders public health efforts. Loop-mediated Isothermal Amplification (LAMP) is an affordable, robust, simple and open detection method which requires only a single incubation temperature and can be done in a cup of hot water. A recent

preliminary report (see Documents section) from New England Biolabs (NEB) used LAMP to detect SARS-CoV-2 in conjunction with a pH indicator to allow visualization of results by a simple color change in 30 minutes or less. If this method can be optimized and compares favorably to the current RT-qPCR test, it will be single-tube nucleic acid amplification method offering rapid, accurate, and cost-effective detection of SARS-CoV-2 that could be deployed anywhere. 

Solution summary in simple terms (CURRENT) (150 words max)

We have developed a simple test for SARS-CoV-2, the causal agent of COVID-19. The test is safe, sensitive and does not require expensive lab equipment. Results can be obtained within one hour of sample collection and can be read via the naked eye as either pink (negative) or yellow (positive).

(Original solution summary in simple terms)

Preliminary data on a LAMP-based method suggests it is sensitive and can be done without complicated equipment, with a simple +/- color change readout, using the commercially-available reagents from NEB. We will optimize and validate a version of this across several labs. This is the first stage of development of a SARS-CoV-2 test that can be run anywhere, even with limited facilities, and by people without a specific background in medicine or biology.   

Solution summary in technical terms (CURRENT) (200 words max)

When this project began in 2020, New England Biolabs (NEB) had recently published preliminary results on a simple SARS-CoV-2 test based on RT-LAMP, a molecular detection method that allows identification of specific nucleic acids in samples with colorimetric visualization of results . The reaction was run for 15-45 min at a single temperature (65 degrees Celsius), and can be done in a cup of hot water with results read by naked eye or cell phone.

We optimized the test protocol by varying the sample collection method, purification and concentration of RNA, primer sequences and ratios, magnesium concentration, sample buffers, and incubation temperature and time. Furthermore, we tested the ruggedness of the method through an interlab study across three labs in the U.S. (West Chicago, IL; Brooklyn, New York; and Atlanta, GA.).

(Original solution summary in technical terms) (200 words max)

New England Biolabs (NEB) recently published preliminary results on a simple SARS-CoV-2 test based on RT-LAMP, a molecular detection method that allows identification of specific nucleic acids in samples with colorimetric visualization of results . The reaction is run for 15-45 min at a single temperature (65 degrees Celsius), and can be done in a cup of hot water with results read by naked eye or cell phone. The price per test is less than 3€. [3] [4] LAMP reagents have been successfully freeze-dried by NEB for detection of other diseases, potentially eliminating the need for a cold chain to deploy it [5].

Initial reports on this test are promising, but the pH -dependent reporter mechanism could potentially be interfered with by sample conditions and the reaction itself is very sensitive to variables such as reaction time. Thus, more work is needed if the test is to become widely accessible and users are of different skill levels.

We would optimize the test protocol by varying primer ratios, magnesium concentration, sample buffers and purification of RNA. Furthermore we would test the ruggedness of the method through an interlab study across at least four labs internationally.

State of advancement of the project (CURRENT) (100 words max)

We have successfully optimized the One Hour COVID-19 Test using LAMP. We have been able to detect both positive and negative samples for the SARS-CoV-2 virus from oral swabs of both symptomatic and asymptomatic individuals. Some of these results have been compared with results of rapid antigen tests and RT-qPCR tests, and results are consistent for RT-qPCR tests, which are the gold standard. Furthermore, we were able to detect a positive sample from an asymptomatic individual that was negative with the rapid antigen test but positive with the RT-qPCR test, which seems to suggest preliminarily that the One Hour COVID-19 Test using LAMP is more sensitive than the rapid antigen test. We have submitted a publication to the Journal of Biomolecular Techniques (JBT) describing our test and preliminary results.

We would now like to apply for Clinical Laboratory Improvement Amendments (CLIA) approval so that we can begin scaled community testing. CLIA approval requires a laboratory director with specific qualifications. We have identified such a laboratory director. In addition, CLIA approved labs must come into compliance with standard operating procedures (SOPs) for clinical testing and with specific requirements for the physical laboratory space.

We would also like to further run clinical validation testing of our One Hour Covid Test directly alongside of RT-qPCR testing. A Methodist hospital system in Indiana is willing to collect samples of suspected COVID-19 positive patients who come to their Emergency Department.

(Original state of advancement of the project)

We have tried at least a dozen different methods of sample collection and subsequent RNA extraction, both those published in preprints and our own iterations of those methods. The majority require centrifugation at at least one step, and those that do not (such as paper-based extraction) were not reproducible in our hands. The protocol that worked best for us leverages three marketed products that are highly unlikely to go scarce. The optimized test takes one hour with only pipettes and a waterbath. 

Project Timeline (CURRENT)

●  Weeks 1 & 2 - Apply for CLIA certification (fill out paperwork and obtain signatures of laboratory owner and laboratory director). Review the requirements for the physical laboratory space and SOPs.  Order a laboratory sink and have it installed in the laboratory. Send oral swab collection kits to the Methodist hospital system in Indiana for clinical validation. Order testing reagents for downstream testing of oral swab samples.

●  Weeks 3 & 4 -  Set up a secure database for patient registration information; to make appointments; to link patients with their sample collection tubes; and to report results.

●  Weeks 5 & 6 - Perform downstream testing of oral swab samples from the Methodist hospital system in Indiana.

●  Weeks 7 & 8 - Begin scaled testing of patients.

(Original Project Timeline)

●  Week 0 - Order reagents (primers, enzymes, buffers, positive control DNA/RNA)

●  Week 1 - Modifications of ratio of primers and magnesium concentrations to assess maximal efficiency of the reaction and its detection. Order synthetic RNA positive controls.

●  Week 2 - Test relating to buffers and treatments for the sample transfer/processing/use using synthetic RNA controls

●  Weeks 3 & 4 - Final Validations of Sensitivity and Specificity with Controls (by day 28)

●  Week 5 - Write up data and transfer methodology to partner lab with BSL2 facilities (CDC partner lab and possibly others)

●  By Week 8 - Conclusions (Is this method robust enough for dealing with samples in this pandemic?) and broad dissemination of the open information.

Note: This was the original timeline, and we spent at least two months exploring many different sample collection and RNA extraction methods. We needed something that was compatible with the colorimetric LAMP kit since we wanted the results to be read our visually with no equipment needed.

Project Implementation

Research: describe hypothesis and research objectives (1000 words max)

We will test the hypothesis that an inexpensive, easy to use kit for detecting SARS-CoV-2 based on isothermal amplification with colorimetric output can be optimized and simplified beyond what is currently commercially available. If the test could be simplified and made more reliable it could allow much wider use by people with minimal training and resources, potentially having a very broad impact. 

Note that as of September 2020 we have coupled the NEB LAMP reagents with the OR-100 swab kit and the prepIT.Q2A reagent from Genotek to develop an open source protocol that can be immediately deployed at scale and used by anyone with a pipettor set and a waterbath (we used a cheap sous vide cooker).

Environmental monitoring would be facilitated for anyone, although use for any patient sample should be run in an appropriate laboratory setting (BSL3). 

The form of isothermal amplification known as Loop-mediated Isothermal Amplification (LAMP) was developed in 2000 (6), and later made even more rapid and able to detect RNA by adding a reverse transcriptase RT-LAMP. LAMP is an established single tube nucleic acid amplification method offering rapid, accurate, and cost-effective diagnosis of infectious diseases (8) at isothermal conditions. It employs a strand displacing DNA polymerase and 4–6 primers at 60-73 degrees Isothermally, and is done in less than an hour (9). The RT-LAMP assay has already been applied for detection of many RNA viruses in humans, such as West Nile, severe acute respiratory syndrome coronavirus, influenza A, mumps, measles, dengue, HIV, and respiratory syncytial virus (10). LAMP can amplify a few copies of the target to 10^9 in less than one hour, even when large amounts of non-target DNA are present or other biological substances (11). RT-LAMP has been shown to be sensitive and specific enough to detect RNA viruses from biological samples such as blood, serum, urine, cerebrospinal fluid, feces, tissue, and nasal swabs, without the need for RNA purification (12). Several recent significant improvements to RT-LAMP have been made, and New England Biolabs (NEB) has recently published a protocol using their WarmStart® Colorimetric LAMP 2X Master Mix (DNA & RNA).

This is a short-term project with several clear objectives:

Objective 1. Validate and optimize NEB’s published protocol. 

We will perform parallel experiments across several international labs, varying conditions and reagents to achieve results that are faster, more reliable and less expensive than the current protocol.

Objective 2. Collect ruggedness data on the optimized protocol

We will ascertain the reliability of the test in the hands of people across several international labs to see if results are consistent.

Objective 3. Test the optimized protocol on inactivated Covid-19 patient samples.

At least one of our participating labs (US Centers for Disease Control) has the requisite biosafety level facilities for testing the improved protocol on patient samples. This is relevant for potential deployment of the test in areas of the world where the machines for the current tests used (WHO and US tests both require a qPCR machine) are scarce.

Methodology (500 words max)

Interlab cooperation and participation is a key part of this methodology. All partner labs will attempt to carry out all of the experiments as their facilities and personnel permit. The objective is to validate the improved protocol across many different labs with varying capabilities and geographic locations.

1. Reproduce the original protocol across several labs. The original NEB publication using RT-LAMP for the detection of SARS-CoV-2 tested five sets of primers. We will start by using the one that worked the best in their hands and for which there is a commercially-available, safe positive control in the form of plasmid DNA. The target is a portion of nucleic acid sequence of the nucleocapsid protein N. The protocol will be as described in Basic NEB LAMP Protocol found in the Document section of this project. 

1. Varying parameters to optimize sensitivity and specificity. We plan to optimize and validate our RT-LAMP primers against two different commercially-available synthetic positive controls (plasmid and RNA) and commercially available controls to check for potential cross-reactivity with MERS and previous SARS. We will measure their sensitivity, specificity, cross-reactivity and TTR (time to response ), and vary primer concentration, incubation times, and magnesium concentrations.

1. Effect of sample buffers. Test samples are collected in various buffers. SInce purification of RNA is often a limiting step in areas with few resources, and since even in developed countries these kits have been in short supply, it is important to determine if testing samples without pre-purification of RNA can yield acceptable results. We will test the effect of different sample buffers on the sensitivity and specificity of the optimized test.

1. Initial validation against current methods using patient samples. This portion of the project will only be carried out by laboratories meeting the biosafety level standards of their country for handling inactivated infected patient samples. Inactivation will be via the DNA Genotek OR-100 kit buffer. For this first phase of the project a limited number of patient samples will be used to compare the results of the optimized test with those obtained through currently used RT-qPCR-based tests.

Results/Expected results (500 words max)

At the end of this phase of the project we will have

1. Developed an optimized version of the RT-LAMP protocol from NEB for SARS-CoV-2 detection.
2. Demonstrated that the protocol can be used across labs of varying resources and by persons with varied levels of previous molecular biology training
3. Documented the process and resulting data that led to 1 & 2 above in a clear and accessible fashion on JOGL
4. Executed preliminary testing of the optimized method on human samples in a partner lab with appropriate biosafety measures.

Safety, quality assurance and regulation

What steps have you taken to ensure your solution’s safety? 

Each participating lab is following biosafety level guidelines in their respective country. Most of the work will be using synthetic positive control samples in a BSL1 area. If the project proceeds to the point where a higher level of biosafety is required (e.g. inactivated human samples) it will proceed only in the participating labs that are equipped to handle samples at that biosafety level. All personnel performing the experiments have received appropriate safety training through their host laboratories to handle the minimal-risk chemical reagents involved in this protocol and use good laboratory hygiene practices.

Note that a key decision was to use the DNA Genotek OR-100 swab kit to collect samples because the buffer in the tube inactivates the virus and renders the samples very safe to handle at the first step of the protocol.

In this first (research only) phase of the project we will not be engaging with the regulatory systems or medical staff. We will not be distributing test kits.

Impact, issues and risks

What impact do you feel your product could have? (100 words max)

There is a shortage SARS-CoV-2 testing throughout the world. New test methods are urgently needed that are simple, affordable and widely accessible. Over-reliance on this complicated technique of molecular amplification limits people in many regions of the globe from knowing whether the virus is present and hinders public health efforts. If we can optimize a simple RT-LAMP method that compares favorably to the current RT-qPCR test, it will offer rapid, accurate, and cost-effective detection of SARS-CoV-2 that could be deployed anywhere.

What do you think would make your project a success? (100 words max)

Success would be completion of the following objectives:

1. Develop an optimized version of the RT-LAMP protocol from NEB for SARS-CoV-2 detection that is sensitive and specific.
2. Demonstrate that the protocol can be used across labs of varying resources and by persons with varied levels of previous molecular biology training.
3. Document the process and resulting data that led to 1 & 2 above in a clear and accessible fashion on JOGL.
4. Execute preliminary testing of the optimized method on human samples in a partner lab with appropriate biosafety measures, showing that it performs as well or better than currently used methods.

Please list the known issues, potential risks, grey-areas, etc. in your project

-Optimization with synthetic controls may result in a test that will not work as well with patient samples

-Some of the reagents are available only from single sources, we need to plan for alternatives such as DIY master mixes. This would also be true if the regents end up unobtainable or costing too much per test to be useful in areas with few resources.

-It may prove impossible to achieve good sensitivity when using crude samples without RNA extraction. If so we may need to segue into alternate purification methods or new sample buffers.

-The test may not be easily shippable despite the reports that these types of reagents can be freeze dried.

Achievement and Benefits of Funding - Results to Date

Updates to this initial project proposal can be found in the Documents tab and also below. There is one project update on 17 May, 2020 and another on 18 October, 2020 (coming soon). 

Optimization of the NEB LAMP Test - Project 163: https://app.jogl.io/project/163

Summary of Work to Date

17 May 2020

This project was funded by JOGL and AXA Research Fund in Round 1. Funds were deposited 23 April, 2020 and therefore week 0 is considered to have begun 23 April. Week 8 would conclude 17 June. However, we have already made progress in all of our proposed milestones within 4 weeks of receiving funding.

Original Round 1 Project Timeline.  

See comments below for an explanation as to what has been completed.

● Week 0 - Order reagents (primers, enzymes, buffers, positive control DNA/RNA)

● Week 1 - Modifications of ratio of primers and magnesium concentrations to assess maximal efficiency of the reaction and its detection. Order synthetic RNA positive controls.

● Week 2 - Test relating to buffers and treatments for the sample transfer/processing/ use using synthetic RNA controls

● Weeks 3 & 4 - Final Validations of Sensitivity and Specificity with Controls (by day 28)

● Week 5 - Write up data and transfer methodology to partner lab with BSL2 facilities (CDC partner lab and possibly others)

● By Week 8 - Conclusions (Is this method robust enough for dealing with samples in this pandemic?) and broad dissemination of the open information.

Evaluation of LAMP Primer Sets

To date, we have evaluated four LAMP primer sets:

1. “Zhang N-A” https://www.medrxiv.org/content/10.1101/2020.02.26.20028373v1

This primer set targets the viral N-gene and is referenced in the above pre-print and is linked directly from the NEB website where the LAMP WarmStart Colorimetric Master Mix can be purchased. The first and last authors work directly for NEB. 

1. “Mammoth N” https://mammoth.bio/wp-content/uploads/2020/03/Mammoth-Biosciences-A-protocol-for-rapid-detection-of-SARS-CoV-2-using-CRISPR-diagnostics-DETECTR.pdf

This primer targets the viral N-gene and is referenced in the above pre-print. The protocol has been developed by Mammoth Biosciences.

3 & 4) “NEB N2” + “NEB E1”. These primer sets are unpublished. They target both the viral N-gene and E-gene. The sequences were provided directly by Nathan Tanner at NEB, who told us they are more sensitive than the Zhang N-A primer set. These primer sets were evaluated separately and together in LAMP reactions. Tanner also recommended adding 40mM guanidine HCl to the LAMP reactions.

Results: We were able to detect SARS-CoV-2 using all four of the above primer sets. However, NEB N2 + NEB E1 together are the most sensitive with an LoD between 10-50 viral copies across three labs (more tests will be run). Positive results are as early as 20 minutes, while negative control samples stay negative for 60 minutes. Optimal conditions were incubation at 68 degrees C (versus the standard 65 degrees C), 2X concentration of the Loop primers and the addition of 40 mM guanidine hydrochloride, pH 8. The Mammoth N-gene primer set performed the worst, with many false positive results for negative controls. The Zhang N-A primer set performed well, but was not as sensitive (LoD ~ 300 viral copies) as the NEB N2 + NEB E1 sets.  

Evaluation of Positive Controls for SARS-CoV-2

After initiating this project, data were published questioning the concentration information provided by the suppliers of some of the controls (e.g. the IDT plasmids). Therefore a variety were tested.

To date, we have evaluated 5 positive controls for SARS-CoV-2:

1. Synthetic N-gene DNA positive control plasmid from IDT: https://www.idtdna.com/pages/landing/coronavirus-research-reagents
2. Synthetic RNA control from Twist: https://www.twistbioscience.com/resources/twist-synthetic-sars-cov-2-rna-controls
3. Heat-inactivated virus in Vero cell culture from Zeptometrix: https://www.zeptometrix.com/products/sars-cov-2-isolate-usa-wa1-2020-culture-fluid-heat-inactivated-05-ml
4. Heat-inactivated virus in human epithelial cell culture from ATCC via BEI Resources: https://www.beiresources.org/Catalog/antigen/NR-52350.aspx
5. Extracted viral RNA from ATCC via BEI Resources: https://www.beiresources.org/Catalog/BEINucleicAcids/NR-52347.aspx

Results: We were able to detect all five of these positive controls for SARS-CoV-2 using the NEB LAMP colorimetric assay. 

Evaluation of Cross-Reactivity for Closely Related Coronaviruses

To date, we have evaluated cross-reactivity for 2 positive controls for other closely related coronaviruses:

1. Synthetic N-gene DNA positive control plasmid for SARS-CoV from IDT: https://www.idtdna.com/pages/landing/coronavirus-research-reagents
2. Synthetic N-gene DNA positive control plasmid for MERS-CoV from IDT: https://www.idtdna.com/pages/landing/coronavirus-research-reagents

Results: We found no cross-reactivity for either the positive control for SARS-CoV or the positive control for MERS-CoV. 

Next Steps:

Evaluation of Sample Collection Methods, Buffers and RNA Concentration/ Purification Methods Using Spiked Synthetic Clinical Samples 

Since this project began, there have been several publications that investigate the number of virus particles in samplings via swab or by collection of fluid such as saliva. Saliva is the easiest fluid to collect, but studies suggest that the number of particles per mL, particularly when diluted in inactivation medium, is too low for direct addition to LAMP to be meaningful in all cases. This is also the case for swabs placed in media such as Viral Transport Media. So, although the LAMP reaction appears to be more rugged than RT-qPCR with regard to interference from sample collection media and the sample components, the most reliable test will probably have to include an RNA extraction step. Saliva is an attractive method to collect samples, because it is non-invasive and there are commercial collection kits that would facilitate home collection. The kit contains inactivation medium and the resulting tube can be mailed safely. The RNA would then be extracted from the tube. Besides purifying the RNA away from potentially interfering components, an extraction step also concentrates the RNA and therefore renders the test more sensitive. 

To date, we have been optimizing the LAMP reactions mostly using purified nucleic acids (synthetic DNA plasmids, synthetic RNA and extracted RNA). We believe we now have optimized primer sets using NEB N2 + NEB E1 and optimized temperature of 68 degrees C, plus 2X concentration of the Loop primers and the addition of 40 mM guanidine hydrochloride. We have begun to explore the creation of synthetic clinical samples. In addition, since existing Covid-19 tests always include a positive control for human RNA, we will begin assessing control primer sets to make sure we can detect human cell RNA as a positive control that our LAMP test is working properly and the sample actually contained material from a patient.

We have begun preliminary tests adding a dilution of the raw Zeptometrix virus in Vero cell culture (not extracted) directly to LAMP reactions and have positive detection. We have also spiked saliva with dilutions of the raw Zeptometrix virus in Vero cell culture, then extracted viral RNA using a commercial kit (E.N.Z.A.), and we have positive detection. 

We have identified several published extraction methods to compare, and are aiming to develop a method that is inexpensive, efficient, and does not depend on backordered reagents. So far we have identified the following methods as promising, and will test them first with spiked synthetic clinical samples.

1. iron bead-based protocol (Hillbilly beads)
2. silica (glass milk) from Harvard paper
3. Syringe filter protocol

Validation Using Actual Clinical Samples in Partnership with Hospitals

-Partners TBD

The final phase of lab work will be to test the performance of the optimized method from sample collection to result using actual clinical samples and compare it to the results using a validated method such as RT-qPCR. This will entail partnerships with access to clinical samples. We are now establishing contacts and have several possible partners, including Chiris Mason at Weill-Cornell Medical Center, the CDC, and international partners.

One Hour Covid Test Using LAMP Test - Project 163: https://app.jogl.io/project/163

(formerly Optimization of the NEB LAMP Test)

Summary of Work to Date

18 October 2020

This project was funded by JOGL and AXA Research Fund in Rounds 1 and 3.  

Goals from Round 3 and results to date.

Our goals from Round 3 funding included 1) Evaluation of sample collection methods, buffers and RNA concentration/ purification methods using spiked synthetic clinical samples; and 2) Validation using actual clinical samples in partnership with hospitals - partners TBD. See Summary of Work to Date 17 May 2020 for more details.

Evaluation of Sample Collection Methods, Buffers and RNA Concentration/ Purification Methods Using Spiked Synthetic Clinical Samples 

Creation and Evaluation of Spiked Synthetic Clinical Samples

Before receiving Round 3 funding, we had begun preliminary creation of spiked synthetic clinical samples using a heat-inactivated culture of SARS-CoV-2 infecting Vero cells from Zeptometrix. Unfortunately, it turned out we could only obtain qualitative results using this culture, because the company did not supply technical information on how many viral genome equivalents per milliliter the culture contained. We therefore ordered another heat-inactivated but quantified culture of SARS-CoV-2 in human lung cells from the American Type Culture Collection (ATCC) via BEI Resources. Use of this actual viral culture has proven important for simulating an actual patient sample when compared to using synthetic or purified viral RNA, because with the actual culture, the capsid must be opened to allow access to the RNA for testing. We consistently find a much higher LOD using the viral culture compared to using the synthetic or purified RNA, which has allowed us to be more realistic in assessment of our test.   

Evaluation of Sample Collection Methods and Buffers

Our team had already been strongly considering the initial use of a commercial kits for sample collection using a kit that includes a buffer that will both inactivate any viral particles present and stabilize the RNA for testing. Contacts were established with sales reps at DNA Genotek, because some of their collection kits had already been approved for FDA Emergency Use Authorization (EUA) and met our criteria for inactivation and stabilization. Two types of sample collection kits were ordered for evaluation: the ORAcollect•RNA OR-100 oral swab kit (https://www.dnagenotek.com/ROW/products/collection-human/oracollect-rna/ORE-100.html) and OMNIgene•ORAL OM-501 saliva kit (https://www.dnagenotek.com/ROW/products/collection-microbiome/omnigene-oral/OM-501.html). We did not have consistent results using the OMNIgene•ORAL OM-501 saliva kit with our downstream testing, so we shifted our focus to the ORAcollect•RNA OR-100 oral swab kit. In addition to getting consistent results with the OR-100, this kit also fills a need we had become aware of when speaking to frontline workers - medical technicians, nurses and physicians - who were asking for a test with an oral swab. The advantages to an oral swab come into play when saliva cannot be collected. An oral swab could be used to test non-responsive patients, patients with esophageal tubes and patients with dry mouths (common in elderly patients). To our knowledge, there is only one test with FDA EUA that uses an oral swab. By using the ORAcollect•RNA OR-100 oral swab kit, we had now decided on both the sample collection method and the buffer that come together.

Regarding buffers, we have also tried many different ones from the myriad of pre-prints that have been released. One concern with using homemade buffers has to do with safety and time. Any new buffer type used should be tested in a third party BSL-3 level lab to make sure it can inactivate viral particles. This would be done by spiking live viral particles into a mammalian cell culture, incubating, and then proving none of the mammalian cells become infected. For the sake of time and safety (and cost), we therefore at this time only wanted to consider buffers that had already undergone this testing. As mentioned above, the DNA Genotek buffer does meet this requirement. One other buffer that meets this requirement is DNA/RNA Shield from Zymo Research. DNA/RNA Shield is sold in collection kits similar to DNA Genotek, but DNA/RNA Shield is also sold by the manufacturer as a stand-alone reagent which provides the advantage of flexibility in collection methods, especially if we eventually move away from premade kits or they become unavailable. This would facilitate more cost-effective testing at point-of-care facilities, in which using the buffer in inexpensive generic collection tubes with separately purchased swabs could easily substitute for the costly OR-100 kits. Unfortunately, results have been inconsistent using Shield with our downstream testing protocol.  

Evaluation of RNA Concentration and Purification Methods

We had previously determined that it is important to purify and concentrate viral RNA. Purification is needed to remove inhibitors such as saliva components and cellular debris, and concentration is needed to avoid false-negative results in low viral titre patients. The purification and concentration step comes between sample collection/ inactivation/ stabilization using the kit and the colorimetric LAMP reaction to yield the result. We spent much of our time in the preceding months evaluating a wide variety of both published protocols and protocol variations we developed. 

In the “Next Steps” section for Round 3 funding, we had proposed trying: 1) an iron bead-based protocol (Hillbilly beads); 2) a protocol using silica (glass milk); and a Syringe filter protocol. The iron bead-based extraction protocol turned out to be costly and there have been shortages of beads, which require precise manufacturing. We switched our experimental focus to silica, paper, syringe filters, alcohol precipitation and other methods. Of the many protocols we tried, silica worked best in our hands. We began by evaluating the glass milk protocol from a Harvard pre-print. However, as with many of the pre-print protocols we tried, we were not able to reproduce their results. We reasoned that silica should work to purify viral RNA because it was commonly used in nucleic acid extraction protocols before column filters and kits became the method of choice. After we tried numerous iterations of modified protocols using glass milk, we found one very simple and inexpensive protocol that worked very well.  

Silica is inexpensive and readily available, making it an attractive component for our test. One drawback of using silica is that a user with minimal training in lab technique might find it difficult to use. Glass milk settles quickly and is granulated and hard to pipette and also because the glass milk protocol requires centrifugation steps; both of these go against scalability, and we felt that to be valuable a test should be scalable. For these reasons we continued to work in parallel on other RNA purification and extraction protocols that did not require glass milk.

We were at the point where we thought that silica-based extraction was our best option when one of our contacts at DNA Genotek suggested that we try the prepIT•Q2A reagent in place of RNA extraction (the purification and concentration step) to see if this reagent might work to shorten the time for our protocol. It worked well in our hands in a modified protocol and have now switched out our glass milk protocol for the prepIT•Q2A protocol. Finally, we tested whether we could eliminate some steps. DNA Genotek recommends pre-heating the sample at 56℃ post-collection, followed by a 10-minute step at 75℃. We did comparisons with and without these steps, and we have determined neither of those are needed. The entire middle step of purifying and concentrating the viral RNA now takes just 20 minutes, which means our entire test from sample collection to result takes just 60 minutes; hence, we changed the name of our project the One Hour Covid Test Using LAMP.

Next Steps:

Validation Using Actual Clinical Samples 

This step turned out to be more difficult than we expected. We pursued several connections but initially were unable to get any clinical lab to commit to evaluating our test. Through Thomas Landrain at JOGL, we now have access to clinical samples to validate our test via the APHP hospital system in Paris. We are in the process of finalizing the logistics for this effort. 

In August it was announced that the XPRIZE foundation was sponsoring a new Rapid Covid Detection challenge. As part of the competition, teams that advance to the semi-finals will be sent ~200 blinded samples for test evaluation. These samples are not clinical samples but are contrived using materials currently being marketed by biotech supply companies as positive controls for Covid-19 testing. Twenty of the teams that pass this first hurdle will be selected for the final round and their tests subjected to a full clinical evaluation using patient samples. We decided to enter the XPRIZE competition to gain publicity for JOGL and our open source test, and more importantly, so that we had a chance to get third party validation of our test. Our team has now advanced to the semi-finals! We are awaiting our contrived samples for validation, and we expect to receive these as early as next week. If validation goes well, we will feel much more confident proceeding to validation using clinical samples.

In addition, we were recently contacted by a group of researchers led by Dr. Mónica Pajuelo at the Molecular Microbiology Laboratory at the Universidad Peruana Cayetano Heredia in Lima, Peru. They said they had read about our One Hour Covid Test in a Medium article written by Marianna Limas with JOGL: https://medium.com/justonegiantlab/a-one-hour-covid-test-developed-by-an-open-source-team-has-advanced-them-to-the-semi-finals-for-127b1781dd86. They are trying to find a reliable alternative to the standard qPCR testing method for Covid-19 in Peru. They have access to clinical samples and are interested in validating our test. We have so far only exchanged emails and have had one Zoom meeting with them, but a collaboration seems promising.

Development of a More DIY Version of the One Hour Covid Test

From the beginning of our project in March of this year, we and the larger collaborative group that communicates via the Nucleic Acid Amplification channel in the JOGL Slack have set an ultimate goal of developing a low-cost, accessible test for the detection of SARS-CoV-2 that avoids as much as possible the use of commercial kits and reagents. As mentioned above, our One Hour Covid Test relies on commercial components from NEB and DNA Genotek that are used together in a novel manner for the first time in a protocol. We made the choice to take this route because our Chicago team member is working to shortly achieve CLIA certification and will be able to offer badly-needed testing capacity in the Chicago area, making a readily available test that is mass manufactured an immediate need. We realize this is not our ultimate end goal, but it serves the immediate and urgent goal of having an open source testing protocol available for use now. Both NEB and DNA Genotek have committed to producing enough of the kits and reagents so that there is not the shortage that exists for so many other kits, kit components and reagents. However, we do still have our ultimate goal in mind, and to that end, we will continue to work towards protocols that do not rely on as many commercial products along with our JOGL partners.

A Big Thank You To JOGL

This project could not have progressed to this stage without the existence of the JOGL platform and without support in the form of mini grants from JOGL and the AXA Research Fund.

List of participating labs and their team members

BioBlaze Community Bio Lab, West Chicago Il, USA.

Sarah Ware has a doctorate degree and has 20 years of research experience as a geneticist/molecular biologist. She is the founder of two independent labs in the Chicagoland area: BioBlaze Community Bio Lab and Lizzy Blossom Ag Services. Sarah also teaches biology and humanities at the university level.

Isabella Zorra has a Masters in Construction Engineering and Management and a Bachelor of Architecture with a focus on fabrication. She currently works as a Project Manager in the Energy Sector. Additionally she is passionate about the sciences and making a lasting improvement on the human condition. Isabella is also with BioBlaze Community Bio Lab.

Lizzy Blossom Ag Services, West Chicago, IL, USA.

Aanika Biosciences, Brooklyn NY, USA.

Ellen Jorgensen has a Ph.D. in molecular biology and has worked in the biotechnology industry for over 30 years, including in FDA-regulated assay development. She has founded two community labs, Genspace and Biotech Without Borders. Her thesis work was on Newcastle Disease virus, a negative strand RNA paramyxovirus. She was a founding staff member of the Sabin Vaccine Foundation, and currently is the Chief Science Officer at Aanika Biosciences, a company offering environmental testing for SARS-CoV-2. https://www.linkedin.com/in/ellenjorgensen/

Hackuarium, Lausanne, Switzerland.

Rachel Aronoff worked on retroviral RNA packaging for her PhD thesis project, cloned an mRNA ‘quality control’ gene from C. elegans, used lentivectors and AAV to study mouse brain circuits, and got implicated in the world of Open Science and Participatory Research when she founded a non-profit - AGiR! Action for Genomic integrity through Research!, but never expected to be involved in anything like this Open Covid19 Initiative. Currently president of the Association Hackuarium and CSO for AGiR!, she loves providing information and promoting research.

Centers for Disease Control, Atlanta GA, USA.

Chris Monaco earned his Masters degree in bioinformatics at the Georgia Institute of Technology. He is currently a microbiologist in the CDC's Division of Scientific Resources' Biotechnology Core Facility Branch (BCFB). https://www.linkedin.com/in/cmonaco/

Center for Research and Interdisciplinarity, Paris, France. 

Guy Aidelberg is completing his PhD on democratizing DNA detection using related techniques and has experience developing similar systems for Zika, Dengue and Chikungunya viruses. He holds a Masters degree in system biology from the Weizmann Institute of Science.

openFIESTA community lab, Madrid, Spain

Francisco Javier Quero: Biologist by training. Tsinghua University Master's degree student and the manager of openFIESTA community lab. Two years working with synthetic biology as the coordinator of Madrid iGEM teams 2018 and 2019. Experience developing isothermal amplifications for infectious diseases. 

Funding 

All partner labs (except the CRI in France; Guy is contributing through protocol design etc. until his institute reopens) are contributing their lab spaces and donating the use of core equipment and person-hours to this project. The funding we are asking for will go towards the purchase of reagents and consumables, and a few pieces of key lab equipment to fill gaps.

Some labs require no funding from JOGL, but those who do have each submitted a separate budget. It is understood that there will be sharing of many reagents as shipping regulations permit.

Please see the individual lab budgets below and in the Documents section (tab above). Funding requests for this round will go towards supplies to create kits for validation on clinical samples and to create more DIY versions of components for our kits.