key: cord-0766682-v59kgcwa
authors: Lownik, Joseph C.; Way, Grayson; Farrar, Jared S.; Martin, Rebecca K.
title: Extraction-free rapid cycle RT-qPCR and extreme RT-PCR for SARS-CoV-2 virus detection
date: 2021-08-25
journal: J Mol Diagn
DOI: 10.1016/j.jmoldx.2021.08.004
sha: 8aa09c9bf4029d42fd14abce4b94ece512e97c18
doc_id: 766682
cord_uid: v59kgcwa

Since the start of the coronavirus disease 2019 (COVID-19) pandemic, molecular diagnostic testing for detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has faced substantial supply chain shortages and noteworthy delays in result reporting after sample collection. Supply chain shortages have been most evident in reagents for RNA extraction and rapid diagnostic testing. This study explored the kinetic limitations of extraction-free rapid cycle RT-qPCR for SARS-CoV-2 virus detection using the commercially available capillary based LightCycler. After optimizing for time and reaction conditions, a protocol for sensitive and specific RT-qPCR of SARS-CoV-2 RNA from nasopharyngeal swabs in less than 20 minutes was developed, with minimal hands-on time requirements. This protocol improves detection speed while maintaining the sensitivity and specificity of hydrolysis probe-based detection. Percentage agreement between the developed assay and previously tested positive patient samples was 97.6% (n= 40/41) and negative patient samples was 100% (40/40). The study further demonstrates that using purified RNA, SARS-CoV-2 testing using extreme RT-PCR and product verification by melting can be completed in less than 3 minutes. Overall, these studies provide a framework for increasing the speed of SARS-CoV-2 and other infectious disease testing.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the virus responsible for coronavirus disease 2019 . Since it was first reported in Wuhan China 1-3 , the virus has spread globally resulting in large scale disruptions to travel and activities, placed extensive pressure on healthcare systems and employees, and caused significant increases in mortality and morbidity. There have been several methodologies implemented for the diagnosis of COVID-19 including traditional reverse transcriptase polymerase chain reaction (RT-PCR) methods 4 , reverse transcriptase loop-mediated isothermal amplification (RT-LAMP) 5, 6 , CRISPR-Cas13 and CRISPR-Cas12 based assays 7-10 as well as serological testing. However, RT-PCR has remained the gold standard for molecular diagnosis of COVID-19.

The current COVID-19 pandemic has shed light into several shortcomings in molecular diagnostic testing. Since the beginning of the pandemic, there have been significant supply chain shortages for supplies from personal protective equipment to molecular diagnostic reagents. Several groups have addressed the shortages of RNA isolation kits by testing different strategies for "extractionfree" SARS-CoV-2 testing. To address supply chain issues revolving around RNA isolation/purification reagents, several groups have demonstrated the feasibility of extraction-free SARS-CoV-2 testing 11-14 . These groups have shown largely concordant results between extraction-free RT-PCR testing compared to RT-PCR using extracted RNA [11] [12] [13] [14] .

In addition to significant supply chain bottlenecks, time-to-results for testing has led to significant delays in diagnosis, necessary isolation/quarantine, as well as treatment. Initial demand during the first wave of COVID-19 in the US resulted in waits of over one week, and during the latest winter wave of infections, time-to-results was 2-4 days for priority patients at large national reference laboratories.

Additionally, hospitals have faced significant shortages of rapid (<1 hour) nucleic acid amplification tests (NAAT) to help triage patients. While RT-PCR tests have been shown to have a very high specificity and sensitivity for SARS-CoV-2, clinical applications of this methodology are traditionally time-consuming and reagent intensive. RT-LAMP has been leveraged in COVID-19 diagnostic platforms, including the J o u r n a l P r e -p r o o f Abbott ID NOW. However, several reports have suggested that this particular test has a high false negative rate in patients with low virus levels [15] [16] [17] .

To address the issues with time-to-results of assays for SARS-CoV-2, previously developed methodologies for extremely rapid RT 18 as well as PCR were utilized 19 . While studies using extreme PCR have focused on intercalating dye-based detection, this study shows that faster thermocycling is possible with hydrolysis probe-based detection by using increased primer concentrations. As a significant bottleneck in supply chain is related to RNA isolation reagents, previously published research regarding extraction-free SARS-CoV-2 testing was utilized [11] [12] [13] [14] which, when paired with rapid cycle RT-qPCR resulted in an assay that can be completed in 20 minutes with little hands-on time. Overall, this study explores the kinetic restraints and optimization of extraction-free RT-PCR SARS-CoV-2 testing and lays the groundwork for future studies to improve the speed of RT-PCR assays.

J o u r n a l P r e -p r o o f

For LightCycler based assays, the SARS-CoV-2 N1 CDC primers and probe were used (Table 1) . Primers for optimal extreme PCR require higher annealing temperatures than what is traditionally used in PCR.

Primers for extreme PCR were designed using the NCBI PrimerBlast suite with optimal annealing temperatures set to 67°C with an oligonucleotide concentration of 10,000 nmol/L and an optimal product melting temperature of 80°C. Primers were then blasted against all virus, bacteria and human Reference Sequences using NCBI PrimerBlast. Primer sequences for extreme PCR can be found in Table 1 .

SARS-CoV-2 genomic RNA, isolate USA-WA1/2020 (BEI # NR-52285), SARS-CoV-2 heat inactivated virus, isolate USA-WA1/2020 (BEI # NR-52286), and SARS-CoV-2 quantitative synthetic RNA (BEI # NR-52358) were obtained from BEI Resources.

For extreme RT-PCR, reactions were performed in 5µL volumes containing 50 mmol/L Tris (pH, 8 

Melting curves were generated using an HR-1 (Idaho Technologies, Salt Lake City, Utah). For experiments comparing relative product amounts, LED voltages were kept constant. High-resolution J o u r n a l P r e -p r o o f melting data were analyzed with custom software written in LabView and viewed as derivative melting curves 20 .

For the LightCycler-based assay, programmed temperatures and times were the following, RT at 55°C for indicated times, followed by 45 cycles of 95 °C for 0 s, and a combined annealing/extension at 63 °C for indicated times. Programmed ramp rates between PCR steps were 20°C/s for all experiments. For extreme PCR, a hot bath (100°C) and a cool water bath (60°C) was used to change sample temperatures, similar to a previously described method 19 . The water baths were heated on electric hotplates with temperature monitoring using an Omega OMB-DAQ-56 USB data acquisition module and Type-T thermocouples (Omega 5SRTC-TT-T-40-36). A stepper motor (stepper online, model #23HS41-1804S), driven by a digital stepper drive (stepper online, DM542T) with pulse and direction signaling provided by an Arduino Uno R3 (Sparkfun) rotated samples in a custom sample holder between each water bath in <0.2s. The stepper motor was controlled using a custom LabView program similar to software previously described 19 . A thermocouple (Omega type T precision fine wire thermocouple, 0.003-inch diameter with Teflon insulation) centered in a dedicated control tube with 5 µL mock PCR mix overlaid with 1 µL mineral oil was used to measure temperature and trigger stepper motion. Stepper motor motion was triggered at 84°C for denaturation and 67°C for annealing/extension to obtain desired temperature cycling profiles for 40 cycles. Reverse transcription was performed for 30 s in the cool bath at 60°C prior to thermocycling.

All reactions were conducted in standard Roche LightCycler capillaries.

10 µL reactions were prepared on ice. Indicated RT times were set on the LightCycler for 55°C followed by a 2-minute 95°C incubation for reverse transcriptase inactivation. Samples then proceeded directly into thermocycling on the LightCycler. Samples were then thermocycled as above with a 30s annealing/extension step. For the 0s RT timepoint, the RT step on the LightCycler was removed and samples were immediately ramped to 95°C at 20°C/s.

J o u r n a l P r e -p r o o f followed by a 2-minute incubation at 95°C. Samples proceeded directly into thermocycling with indicated annealing/extension times. For the 0s timepoint, the hold time for annealing/extension at 63°C was set to 0s.

Nasopharyngeal swabs were collected in 1mL of universal transport medium (UTM; BD). 

While it has been shown that reverse transcription (RT) can be performed faster than manufacturer's recommend time 18 , we wanted to determine the shortest amount of time required for efficient RT. While extreme PCR and extreme RT typically use primer concentrations in the range of 5-20 μmol/L 18,19 , 5 μmol/L primer concentrations were used as capillary based LightCycler PCR cycle times (20-40s) are much slower than those used in extreme PCR (<1.2s). The longer cycle times on the LightCycler would not require increased polymerase concentrations that might necessitate further increases in primer concentration. Luna Warm Start Reverse Transcriptase (NEB) was utilized because of its aptamer based warm start inhibition, the reversibility of this warm start formulation, as well as its increased thermostability. As a proof of principle to further generalize the results, the CDC SARS-CoV-2 assay N1 primer and probe set was used.

Previous studies examining rapid RT primarily utilized isolated RNA 18 . To optimize this assay for extraction free testing, the time requirements for RT were examined with pure RNA, SARS-CoV-2 genomic RNA was spiked into normal saline, as well as 10 samples for which the patients had tested positive (prepared as in methods). After setting up reactions on ice, the reaction tubes were incubated at 55°C for 0-600s. Reactions were then subjected to PCR amplification using the capillary based LightCycler 1.5 with 30s annealing/extension times. Similar to previous results 18 , efficient RT within 30s for multiple concentrations of pure SARS-CoV-2 RNA was observed ( Figure 1A) . However, extraction-free positive patient samples required between 5-10 minutes for efficient RT (Figures 1B-C) . Next, whether increasing RT enzyme concentration was able to affect the time required for efficient RT was tested using both pure SARS-CoV-2 RNA as well as extraction-free positive patient samples. Higher RT enzyme concentrations further reduced the time required to perform efficient RT (Figures 1D-E) . However, important differences were observed when using extraction-free positive patient samples for the optimal time and RT enzyme concentrations required for efficient RT-qPCR ( Figure 1E ). With 1X RT enzyme concentration, >5 J o u r n a l P r e -p r o o f minutes was required for efficient RT-qPCR when using extraction-free patient samples, however, when 4X RT enzyme was used, this time could be reduced to 2 minutes for efficient RT-qPCR. These results demonstrate important differential kinetic limits for RT when using purified RNA versus extraction-free patient samples.

As previous studies have shown an inhibitory effect of extraction-free sample preparation on the PCR phase of SARS-CoV-2 RT-qPCR, we wanted to determine an optimal PCR protocol for our assay.

Most commercial RT-PCR assays for SARS-CoV-2 have an annealing/extension time of ≥ 30s per cycle.

Annealing/extensions times of < 0.5s per cycle for short PCR products can result in efficient and specific amplification 19, 21 . However, these previous studies using extreme PCR were conducted on purified DNA samples. Additionally, the CDC SARS-CoV-2 assay chosen uses hydrolysis probe-based detection, whereas extreme PCR has used intercalating dye chemistry for detection. Again, purified RNA or extraction-free positive patient samples were used to test for the optimal time required for this assay.

Samples were reverse transcribed for 10 minutes prior to thermocycling.

Surprisingly, there was only a quantification cycle (Cq) difference of approximately 1 cycle for purified SARS-CoV-2 RNA when the annealing/extension step was set for 0s versus 30s, suggesting efficient PCR can be done with very rapid annealing/extension times (Figure 2A) . However, extractionfree positive patient samples required a 10s or greater annealing/elongation step for efficient PCR ( Figure   2B ) and this time could not be reduced by increasing Taq polymerase concentrations in the reaction (Figures 2C-D) . Using a 10s annealing/elongation step, the reaction efficiency was examined using pure RNA as well as RNA spiked into diluent from an uninfected control NP swab. Both the pure RNA and spiked NP sample were efficient with efficiencies of 100.3% and 109.9%, respectively (Figure 2F-G) . Additionally, 20/20 reactions with 3 copies/reaction tested positive in spiked NP samples suggesting adequate detection of 2,000 copies/mL when dry NP swabbing is performed for testing ( Figure 2H ).

Based on our optimization studies, an RT-qPCR protocol which consisted of a 5-minute RT step at 55°C followed by a 10s denaturation step at 95°C was adopted. This was then followed by 45 cycles of PCR with a 10s annealing/elongation at 63°C with a 0s hold time at 95°C. Overall, this PCR protocol takes a total of 20 minutes (Figure 3A) . 41 positive patient samples were tested using extraction-free testing. 16 

While the LightCycler protocol allowed for successful detection of SARS-Cov-2 in 20 minutes using extraction-free methodologies, we wanted to examine whether the overall time for testing could be decreased using RNA extraction followed by extreme RT-PCR (Figure 4A) . For extreme RT-PCR, a different primer set targeting the nucleocapsid gene of SARS-CoV-2 was utilized. Successful amplification was measured by high resolution melting curve analysis of the reaction product. Using a 30s RT step at 60°C followed by 40 PCR cycles (~1.4 s/cycle), purified SARS-CoV-2 genomic RNA was successfully amplified with detection down to 4 copies/reaction ( Figure 4B) . Additionally, amplification of 58 copies of SARS-CoV-2 genomic RNA spiked into NP swab diluent from a healthy, uninfected control was observed ( Figure 4C) . However, extraction free SARS-CoV-2 positive patient samples were not successfully amplified (data not shown), likely due to both RT and PCR inhibition observed in these samples (Figures 1 and 2) . Overall, extreme RT-PCR followed by high resolution melting required < 3 minutes to complete.

The COVID-19 pandemic has made it abundantly clear that time-to-results for molecular diagnostic testing for pathogens is very important in not only making clinical decisions, but also for identifying individuals that should quarantine to prevent further community transmission. Additionally, significant amounts of resources and time are required to keep patients in emergency departments and hospitals isolated until results from pathogen testing come back, sometimes more than 2 days later.

However, it has also become clear that the sensitivity of an assay cannot be traded for speed, as false negative results can have detrimental and life-threatening impacts to patients, healthcare workers, and the community.

While many new advances in molecular diagnostic testing have been developed in response to the COVID-19 pandemic, shortages in rapid testing supplies have remained. Several groups have shown the feasibility of extraction-free RT-PCR for SARS-CoV-2. This methodology not only decreases time requirements of testing but also allows for more economical testing. In addition to the time requirements of RNA isolation, typical plate-based thermal cycler RT-PCR reactions take >60 minutes putting the entire process of RNA isolation and RT-PCR at several hours.

We chose to examine the kinetic constraints of extraction-free RT-PCR for SARS-CoV-2 RNA detection to facilitate the creation of an economical and rapid molecular diagnostic test for COVID-19.

While extreme and rapid PCR are typically conducted without a probe 19, 22 , we chose to utilize probe based methodology for increased specificity. Additionally, the LightCycler (Roche) RT-PCR instrument was used for thermocycling as it is able to rapidly thermocycle allowing for better optimization studies 23 .

However, our methodology was also applied to a plate-based instrument (QuantStudio 3, ThermoFisher) with similar results, albeit a longer overall time of ~30 minutes due to slower temperature ramp rates (data not shown), showing the broader adaptability of our methodology. Additionally, using the QuantStudio 3, we were able to multiplex the CDC N1 primer set with the CDC RPP30 primer set with concordant results to our LightCycler assay (data not shown). These results suggest that both the RT (Figure 1) and PCR steps (Figure 2 ) are inhibited during extraction-free RT-PCR. Interestingly, J o u r n a l P r e -p r o o f differential inhibition between patient samples was observed with a more striking difference in inhibition between uninfected spiked samples and positive samples for both RT and PCR. This suggests a possibility of increased inhibition with increased inflammation/mucus during infection. This further confirms that optimization studies require the use of samples from infected patients rather than uninfected spiked samples.

Our data (Figures 1B-D) shows that increasing Taq concentration in extraction-free RT-PCR does not decrease the required anneal/extension time, suggesting that the inhibition is not polymerase concentration dependent, but rather the inhibition is more likely affecting the processivity rate of the Taq polymerase. Additional experiments would be needed to parse out whether the inhibition seen in extraction-free RT-PCR primarily affects the polymerase activity or 5' 3' exonuclease activities of Taq.

We also demonstrate that an increase in primer concentration (5 µM) is required for the rapid thermocycling (≤ 10s) in our assay. While our primer concentration was increased ~10-fold above traditional primer concentrations, our assay required only a modest 2-fold increase in probe concentration (250 nM). Future studies will be needed to determine whether optimization of primer and probe concentrations can additionally increase the speed of our protocol.

Using a total of 20% sample in the reaction, extraction-free RT-PCR requires ~5-10 minutes for efficient RT and 10s annealing/elongation per cycle for PCR which results in a total time of 20 minutes for a 45-cycle reaction using a LightCycler 1.5 (Roche). However, using isolated RNA, RT-PCR can be shortened to < 30s for RT and < 60s for 45 cycles of PCR for a total reaction time of ~90s. These results suggest that both methods can allow for rapid testing but the limit for extraction-free seems to be ~20 minutes. The difference in time requirements for these two assays (~18 minutes) suggests that if a rapid RNA isolation step is implemented, the time required may be decreased even further. Pairing together microfluidic and rapid thermocycling platforms may allow for rapid RNA isolation and extreme RT-PCR thermocycling for an even faster result. Another potential way to decrease inhibition would be to use a smaller amount of patient sample (< 20%), however, this may decrease overall sensitivity of the assay.

J o u r n a l P r e -p r o o f Additionally, it may be possible to increase the speed of our extraction-free protocol using polymerases and/or reverse transcriptases which are known to have higher inhibitor resistance 24, 25 .

Additional time could be saved by collecting samples directly into a low-complexity buffer (saline or molecular grade water) containing Triton X-100. Additionally, future clinical studies examining the differential inhibition of nasopharyngeal, oropharyngeal and sputum samples is warranted to further refine optimal sample collection for our protocol and other extraction-free methods. While our study did not directly examine the role of different detergents and/or heating strategies for extraction-free RT-PCR, optimization with Tween-20, NP-40 or other detergents in combination with a rapid heating step may further allow for more rapid RT and PCR. Our results showing successful amplification of 58 copies of SARS-CoV-2 using extreme RT-PCR but not in the presence of matrix from positive patients suggests a different and/or increased inhibitory mechanism in positive patients vs. healthy controls. Future studies will be needed to elucidate this mechanism as well as ways to address this inhibition for extraction-free extreme RT-PCR to be feasible.

While testing of the protocol developed in these studies was limited to the use of CDC N1 primers and extreme RT-PCR primers residing in the N gene of SARS-CoV-2, our results lay the groundwork for a rapid and economical molecular diagnostic assay which still maintains the sensitivity and specificity required for accurate diagnoses. This study demonstrates the feasibility of swab to result RT-PCR using an in vitro diagnostic approved LightCycler (Roche) paired with commercially available RT-PCR reagents. Both methods showed sensitivity and specificity required for screening patients for SARS-CoV-2. 

A novel coronavirus from patients with pneumonia in China

A new coronavirus associated with human respiratory disease in China

RT-QPCR testing of SARS-COV-2: A primer

Development of Reverse Transcription Loop-Mediated Isothermal Amplification Assays Targeting Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2)

RT-LAMP for rapid diagnosis of coronavirus SARS-CoV-2

CRISPR-based surveillance for COVID-19 using genomically-comprehensive machine learning design

CRISPR-Cas12-based detection of SARS-CoV-2

Ultrasensitive and visual J o u r n a l P r e -p r o o f detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay

Amplification-free detection of SARS-CoV-2 with CRISPR-Cas13a and mobile phone microscopy

An alternative workflow for molecular detection of SARS-CoV-2 -escape from the NA extraction kit-shortage

Detection of SARS-CoV-2 RNA by direct RT-qPCR on nasopharyngeal specimens without extraction of viral RNA

Massive and rapid COVID-19 testing is feasible by extraction-free SARS-CoV-2 RT-qPCR

Rapid and extraction-free detection of SARS-CoV-2 from saliva with colorimetric LAMP

Comparison of Four Molecular In Vitro Diagnostic Assays for the Detection of SARS-CoV-2 in Nasopharyngeal Specimens

Clinical Evaluation of Three Sample-To-Answer academic institution

Reverse transcriptase kinetics for one-step RT-PCR

Extreme PCR: Efficient and Specific DNA Amplification

Chapter 13 Mathematical Algorithms for High-Resolution DNA Melting Analysis

Influence of PCR Reagents on DNA Polymerase Extension Rates Measured on Real-Time PCR Instruments

Microfluidic Genotyping by Rapid Serial PCR and High-Speed Melting Analysis

LightCycler technology in molecular diagnostics

Direct DNA amplification from crude clinical samples using a PCR enhancer cocktail and novel mutants of Taq

Mutants of Taq DNA polymerase resistant to PCR inhibitors allow DNA amplification from whole blood and crude soil samples