key: cord-0987309-pxzxaoup
authors: Yang, J.; Salfati, E.; Kao, D.; Mihaylova, Y.
title: Use of alternative RNA storage and extraction reagents and development of a hybrid PCR-based method for SARS-CoV-2 detection
date: 2020-11-24
journal: nan
DOI: 10.1101/2020.11.21.20236216
sha: a42c796206eb49434914964a173ac12d7d1494ef
doc_id: 987309
cord_uid: pxzxaoup

The COVID-19 pandemic has presented multiple healthcare challenges, one of which is adequately meeting the need for large-scale diagnostic testing. The most commonly used assays for detection of SARS-CoV-2, including those recommended by the Center for Disease Control and Prevention (CDC), rely on a consistent set of core reagents. This has put a serious strain on the reagent supply chain, resulting in insufficient testing. It has also led to restricted animal testing, even though there are now multiple reports of animals, particularly cats, ferrets and minks, contracting the disease. We aimed to address the diagnostic bottleneck by developing a PCR-based SARS-CoV-2 detection assay for cats (and, potentially, other animals) which avoids the use of most common reagents, such as collection kits optimized for RNA stabilization, RNA isolation kits and TaqMan-based RT-PCR reagents. We demonstrated that an inexpensive solid-phase reversible immobilization (SPRI) method can be used for RNA extraction from feline samples collected with the ORAcollect RNA OR-100 and PERFORMAgene DNA PG-100 sample collection kits (available from DNAGenotek), optimized for RNA or DNA stabilization, respectively. We developed a dual method SARS-CoV-2 detection assay relying on SYBR RT-PCR and Sanger sequencing, using the same set of custom synthesized oligo primers. We validated the specificity of our test with a commercially available SARS-CoV-2 plasmid positive control, as well as two in-house positive control RNA samples. The sensitivity of our assay was determined to be 10 viral copies per reaction. Our results suggest that a simple SPRI-dependent RNA extraction protocol and certain sample collection kits not specifically optimized for RNA stabilization could potentially be used in cases where reagent shortages are hindering adequate COVID-19 testing. These alternative reagents could be used in combination with our COVID-19 testing method, which relies on inexpensive and readily available SYBR RT-PCR and non-fluorescent PCR reagents. Depending on the detection goals and the laboratory setup available, the SYBR RT-PCR method and the Sanger sequencing based method can be used alone or in conjunction, for improved accuracy. Although the test is intended for animal use, it is, in theory, possible to use it with human samples, especially those with higher viral loads.

CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint / 69 Introduction 70 The COVID-19 pandemic has put an unprecedented strain on almost every country's healthcare 71 system. In addition to insufficient hospital beds and personal protective equipment for 72 healthcare workers, this pandemic has also been marked by a shortage of COVID-19 diagnostic 73 testing. The most frequently used assay type, the SARS-CoV-2 RT-PCR, relies on the same set 74 of preferred core reagents. For this type of test, the CDC recommends the use of commercially 75 available RNA isolation kits and RT-PCR reagents designed for use with an oligo probe (i.e., the 76 TaqMan approach) 1 . This has led to multiple reports of RNA isolation kit shortages 2 , slowing 77 down testing efforts at a time when testing speed is crucial. In addition, from our own internal 78 observations, many TaqMan-based RT-PCR reagents, as well as nasopharyngeal and 79 oropharyngeal swab kits optimized for RNA stabilization, were backordered and/or experienced 80 a price increase at some point during the pandemic. These supply chain bottlenecks have also 81 led to very sparse testing of animals suspected to have COVID-19, despite the fact that there 82 has been strong research evidence indicating that animals (particularly cats, ferrets and minks) 83 can contract and spread COVID-19 to other animals 3,4,5 . Even with limited testing, there are 84 currently over 60 reported COVID-19 positive animal cases in the US alone, as catalogued by 85 the United States Department of Agriculture 6 . Recently, COVID-19 diagnostic tests designed 86 specifically for animal use were made available for clinical use only 7 . However, these tests still 87 rely on the same set of core reagents and, therefore, do little to address or circumvent reagent 88 shortages. To address this issue, we developed a SARS-CoV-2 PCR-based test for use in cats 89 (and potentially other animals) that does not rely on the reagents most typically used in 90 SARS-CoV-2 RT-PCR tests. Our assay can be used with a sample collection kit optimized for 91 DNA stabilization or RNA stabilization and does not use commercial RNA extraction kits or 92 TaqMan based RT-PCR reagents. Our test is comprised of two different workflows -SYBR 93 RT-PCR and Sanger sequencing. These can be used together or separately, depending on the 94 available reagents and lab setup. The availability of such a test would be important for 95 improving our understanding of the spread and impact of the SARS-CoV-2 virus in animals. 101 We collected an oropharyngeal swab sample from the same cat using two different swab 102 collection kits -the ORAcollect RNA OR-100 and the PERFORMAgene DNA PG-100 (both 3 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint / 103 available from DNAGenotek), optimized for RNA and DNA stabilization, respectively. We then 104 extracted total RNA from each of these samples using a published method for purifying nucleic 105 acids by solid-phase reversible immobilization (SPRI) 8 , which does not rely on a commercial kit. 106 Gel electrophoresis indicated that we were successful in extracting total RNA from each of the 107 two samples ( Figure 1A ). Quantification of the extracted RNA suggested that our extraction 108 protocol yielded higher quantities of RNA from the sample collected with the PERFORMAgene 109 DNA PG-100, compared to the sample collected with the ORAcollect RNA OR-100 ( Figure 1B ).

110 RNA purity was quantified by absorbance measurements taken at 260/280 nm and 260/230 nm. 111 Our data suggest that the RNA extracted from the ORAcollect RNA OR-100, although lower 112 quantity, is of better quality. However, the best indicator of RNA quality is, ultimately, its 113 functionality in a particular application of interest. We wanted to investigate whether the RNA 114 extracted from the PERFORMAgene DNA PG-100 sample contained feline mRNA and viral 115 RNA. We successfully detected two housekeeping genes (GAPDH and RSP19) in each of the 116 two samples, although RSP19 was of a lower quantity in the RNA extracted from the 117 PERFORMAgene DNA PG-100 sample compared to the RNA extracted from the ORAcollect 118 RNA OR-100 sample ( Figure 1C ). Therefore, while both collection kits allowed successful 119 extraction of feline mRNA from an oropharyngeal sample, our results indicate that some feline 120 mRNAs might have different abundance between the two collection kits. 121 122 We next tested whether the total RNA extracted from the PERFORMAgene DNA PG-100 123 sample contained viral sequences. We prepared an RNA-seq library from the extracted RNA 124 and ran the sequencing results through a microbial sequence classifier to identify 33 viral 125 genomes, at least 9 of which had RNA genomes ( Table 1 ) . Surprisingly, despite the DNAse 126 treatment step in the RNA-seq library preparation process, we still detected some viruses with 127 DNA genomes in the sample. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint / 136 In an attempt to avoid using TaqMan RT-PCR reagents for SARS-CoV-2 detection, we tested 137 whether a combination of SYBR RT-PCR and Sanger sequencing could deliver reliable 138 detection of the virus. We designed oligo primers amplifying a 328 bp sequence in the N gene 139 region of the SARS-CoV-2 genome and looked at the conservation between the homologous 140 target regions of SARS-CoV-2, SARS-CoV and the Feline enteric coronavirus (FECV) ( Figure   141 2 ). The FECV sequence is sufficiently divergent as to not present a threat for viral 142 cross-detection. SARS-CoV, however, shares high homology with SARS-CoV-2 and our primer 143 pair would work equally well for amplifying either of the two viral sequences. Therefore, the 144 SYBR RT-PCR part of our assay, when used in isolation, cannot distinguish between the two 145 viruses. The Sanger sequencing component of our test, on the other hand, can clearly 146 distinguish between SARS-CoV and SARS-CoV-2 due to the single nucleotide resolution that it 147 provides ( Figure 3 ). Figure 4B ). Next, we examined our test's sensitivity using SYBR RT-PCR 157 ( Figure 4C ). Results showed that our assay's limit for reliable detection is 10 viral copies per 158 reaction. We used a Cq value of 35 cycles as the cut off for reliable detection. We observed that 159 we could not reliably and reproducibly detect 1 or fewer viral copies per reaction. This suggests 160 that our test would work best on samples with higher viral loads and might, in theory, miss cases 161 where the disease is at its initial stage and the viral load is still relatively low. 162 163 We next tested our assay with two in-house RNA positive control samples obtained from an 164 anonymous source. Each of them had a known concentration of 100 viral copies per reaction. 165 We transcribed cDNA from these samples and ran our conventional PCR assay in order to 166 perform Sanger sequencing on the resultant products. We observed a 100% concordance 167 between the SARS-CoV-2 positive samples' viral sequences and the published SARS-CoV-2 168 viral genome sequence ( Figure 5 ). While preliminary, these results suggest that our assay 5 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint / 169 could be used in a real-world scenario to identify SARS-CoV-2 positive feline and, potentially, 170 human patients. 198 199 Our test's sensitivity is 10 viral copies per reaction. Human patients' SARS-CoV-2 viral loads in 200 nasopharyngeal and oropharyngeal samples typically vary between 1.9 and 8 log 10 RNA 201 copies/ml 9 , which corresponds to a range between 0.395 and >5000 viral copies per reaction 202 volume as defined in our assay. While not much data is available on the comparative viral loads 6 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint 7 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint / 237 238 cDNA synthesis 239 We used the iScript Select cDNA Synthesis Kit (1708896) from BioRad and followed 240 manufacturer's instructions. 

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint

COVID-19 positive animal patients, these results suggest that our test is likely more suitable

Although more studies are needed, our test could, in theory

This 210 will allow a more accurate strain-level classification of positive cases without resorting to the 211 much slower and more expensive whole viral genome sequencing. Using our method

While neither kit is designed specifically for 220 oropharyngeal sample collection, we found it relatively straightforward to use either of them to 221 collect an oropharyngeal swab sample from a cat. Samples were collected in the same sitting, 222 from the same cat, with both collection kits. The cat had not had anything to eat or drink for an (SPRI) 8,10 , with some protocol variations

First, each sample was heated at 55°C for 1 hour. The sample collected with the ORAcollect 229 RNA OR-100 kit was then heated for 10 min at 90°C and the pH was adjusted to neutral, as per 230 manufacturer's instructions

PERFORMAgene DNA PG-100. Next, 2X volume of SPRI beads mix (Magnetic Beads

200 from MCLAB) was added to each sample, mixed and incubated for 5 min 233 prior to bead immobilization with a magnet. The bead pellet was washed twice with 80% freshly 234 prepared ethanol and the RNA was eluted in 1XTE

Sodium acetate (pH 5.5) and 100% ethanol were used for RNA precipitation for 30 236 min at -20°C. Samples were then pelleted, washed with 70% ethanol and resuspended in 1XTE

Figure 1. Yield and quality of RNA extracted from a feline oropharyngeal swab sample 339 collected with the ORAcollect RNA OR-100 (RNA swab) and the PERFORMAgene DNA 340 PG-100 (DNA swab). (A) Gel electrophoresis comparing the total RNA extracted from each 341 sample collection kit. The M lane contains a nucleic acid size marker. (B) Measures of RNA 342 amount and purity using ThermoFisher Scientific NanoDrop and Promega Quantus 343 spectrophotometers. (C) RT-PCR results quantifying the expression of two housekeeping genes 344

Table 1. High, medium and low abundance viral sequences detected in the RNA extracted 347 from the PERFORMAgene DNA PG-100 sample. Viruses with known RNA based genomes 348 are marked in orange

SARS-CoV and FECV viral 351 genomes, focusing on the corresponding region targeted for amplification in our assay. 352 (A) Nucleotide alignment between the 3 viruses, with the primers used in our assay shown in 353 color; red indicates nucleotides that are the same between all three viruses and green indicates 354 nucleotides shared only between SARS-CoV-2 and SARS-CoV. (B) Protein alignment of the 355 whole 328 bp region targeted for amplification in our assay

Top 3 results when (A) the SARS-CoV-2 or (B) the SARS-CoV region targeted for 358 amplification by our primers is used as input into NCBI's BLASTp tool

A) Gel electrophoresis with PCR results from titrated 362 positive and negative control samples using a primer pair amplifying a 328 bp region of the 363 SARS-CoV-2 genome. (B) Nucleotide alignment between the published N gene SARS-CoV-2 364 sequence (Query) and a Sanger sequenced PCR product from the experiment in (A) marked as 365 (Sbjct). (C) SYBR RT-PCR testing the sensitivity of the assay to different SARS-CoV-2 viral 366 loads. Feline cDNA was spiked-in with titrated amounts of SARS-CoV-2 positive control 367 plasmid

Nucleotide alignment between the published N gene SARS-CoV-2 sequence 370 (Sbjct) and Sanger sequenced PCR products generated with our assay's primers using 371 two SARS-CoV-2 RNA positive control samples (A,B). The Sanger sequenced PCR products 372 are marked as

CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprintThe copyright holder for this this version posted November 24, 2020. ; https://doi.org/10.1101/2020.11.21.20236216 doi: medRxiv preprint