key: cord-0756713-xwzmeijf
authors: Mondal, S.; Feirer, N.; Brockman, M.; Preston, M.; Teter, S.; Ma, D.; Goueli, S. A.; Moorji, S.; Saul, B.; Cali, J. J.
title: A Direct Capture Method for Purification and Detection of Viral Nucleic Acid Enables Epidemiological Surveillance of SARS-CoV-2
date: 2021-05-10
journal: nan
DOI: 10.1101/2021.05.06.21256753
sha: d9aa573fc2ad239395216532d49805272b872254
doc_id: 756713
cord_uid: xwzmeijf

Studies have demonstrated that SARS-CoV-2 RNA can be detected in the feces of infected individuals. This finding spurred investigation into using wastewater-based epidemiology (WBE) to monitor SARS-CoV-2 RNA and track the appearance and spread of COVID-19 in communities. SARS-CoV-2 is present at low levels in wastewater, making sample concentration a prerequisite for sensitive detection and utility in WBE. Whereas common methods for isolating viral genetic material are biased toward intact virus isolation, it is likely that a relatively low percentage of the total SARS-CoV-2 RNA genome in wastewater is contained within intact virions. Therefore, we hypothesized that a direct unbiased total nucleic acid extraction method could overcome the cumbersome protocols, variability and low recovery rates associated with the former methods. This led to development of a simple, rapid, and modular alternative to existing purification methods. In an initial concentration step, chaotropic agents are added to raw sewage allowing binding of nucleic acid from free nucleoprotein complexes, partially intact, and intact virions to a silica matrix. The eluted nucleic acid is then purified using manual or semi-automated methods. RT-qPCR enzyme mixes were formulated that demonstrate substantial inhibitor resistance. In addition, multiplexed probe-based RT-qPCR assays detecting the N1, N2 (nucleocapsid) and E (envelope) gene fragments of SARS-CoV-2 were developed. The RT-qPCR assays also contain primers and probes to detect Pepper Mild Mottle Virus (PMMoV), a fecal indicator RNA virus present in wastewater, and an exogenous control RNA to measure effects of RT-qPCR inhibitors. Using this workflow, we monitored wastewater samples from three wastewater treatment plants (WWTP) in Dane County, Wisconsin. We also successfully sequenced a subset of samples to ensure compatibility with a SARS-CoV-2 amplicon panel and demonstrated the potential for SARS-CoV-2 variant detection. Data obtained here underscore the potential for wastewater surveillance of SARS-CoV-2 and other infectious agents in communities.

. 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

In late December of 2019, Chinese health authorities examined a new respiratory virus that caused 72 unexplained cases of severe pneumonia 1 . Subsequent sequencing identified the virus as a member of the 73

Coronavirus family, a group of enveloped RNA viruses that commonly infect birds, mammals, and 74 humans 2 . The novel virus, designated Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), 75 is related to SARS-CoV and MERS-CoV, other respiratory viruses that can lead to fatal illness 3 . SARS-76

CoV-2 was determined to be the causative agent of the respiratory disease COVID-19 1,4 . COVID-19 77 quickly evolved into a global pandemic, which at the time of this report has resulted in over 3.1 million 78 deaths worldwide (https://coronavirus.jhu.edu/map.html). 79

The rapid person-to-person spread of SARS-CoV-2 is in part due to the high infectiousness of viral 81 carriers 5,6 . Pre-symptomatic shedding is thought to drive a significant amount of viral spread, as the 82 highest risk of transmission occurs very early in the course of the disease 5,6 . Asymptomatic carriers of the 83 virus are also infectious and display similar viral loads in the respiratory system, despite exhibiting faster 84 viral clearance 6-9 . As traditional nasal swab testing approaches can be biased towards symptomatic viral 85 carriers 10 , viral surveillance methods that provide a widespread view of community infection are vital for 86 accurate monitoring and control of the ongoing pandemic. 87

Although infectious SARS-CoV-2 virions are rarely isolated from feces 11 , several reports early in the 89 COVID-19 pandemic demonstrated that SARS-CoV-2 can frequently (50%-70%) be detected in fecal 90 samples from both symptomatic and asymptomatic infected individuals 9,12-15 . Levels of SARS-CoV-2 91 RNA in feces are not correlated with the presence/absence of gastrointestinal illness or with overall 92 COVID-19 disease severity 12 , as is often the case with respiratory samples. Although SARS-CoV-2 viral 93 load is consistently higher in respiratory specimens, viral RNA in feces can be detected significantly 94 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint 6 FR3710; Sun Prairie: ISCO 5800; Oregon: ISCO 4700). Samples were kept at 4°C at all times during 146 transport and storage. All samples were processed within six hours of sampling. Total nucleic acid (TNA) was purified from collected wastewater using the Wizard ® Enviro Wastewater 157 TNA kit (Promega Corp.) and/or the Maxwell ® Enviro Wastewater TNA kit (Promega Corp.), both of 158 which use an initial concentration step composed of direct capture of nucleic acids on silica resin. Briefly, 159 0.5 mL of alkaline protease was added to 40 mL of untreated wastewater in triplicate, and samples were 160 incubated statically for 30 minutes at room temperature. The samples were then centrifuged at 3000 x g 161 for 10 minutes to remove suspended solids. The supernatant was transferred to a new vessel and 12 mL of 162

Binding Buffer 1 and 1 mL of Binding Buffer 2 were added, followed by gentle mixing. A 48 mL of 163 isopropanol was added to the mixture, gently mixed and passed through a PureYield™ Midi Binding 164

Column (Promega Corp.) using a VacMan ® Vacuum Manifold (Promega Corp.). Nucleic acid captured on 165 the PureYield™ Midi Binding Column was washed with 5 mL of Column Wash 1 followed by 20 mL of 166 Column Wash 2. Nucleic acid was eluted with nuclease-free water (Fig. 2) . 167

The eluted nucleic acid was further purified using a Mini spin column for the Wizard ® Enviro Wastewater 169 TNA kit or with an automated nucleic acid purification system (Maxwell ® RSC, Promega Corp) for the 170 . 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint 400 µL of Binding Buffer 1, 100 µL of Binding Buffer 2 and 1.5 mL of isopropanol are added to the 1 172 mL of nucleic acid extracted in the concentration step, and then passed through a spin column with a 173 silica resin. The column is washed with 350µL and 1 ml of Column Wash 1 and 2 respectively, and 174 nucleic acid is extracted in 80 µL of water. For the automated Maxwell ® Enviro Wastewater TNA kit , 175 150 µL of Binding Buffer 1 and 50 µL of Binding Buffer 2 are added to 0.5 mL of nucleic acid extracted 176 in the concentration step. Then the total volume is added to well #1 of the Maxwell ® Cartridge and 177 nucleic acid is eluted in 80 µL of nuclease-free water. 178 179 180 Isolation of Total Nucleic Acid from suspended solids using the direct capture method 181

The method to extract total nucleic acids from suspended solids is a variation on the direct capture 182 methodology described above. After the alkaline protease step, the pellet of suspended solids is 183 resuspended in 5 mL of nuclease-free water. To the resuspended pellet, 1.5 mL of Binding Buffer 1, 125 184 μL of Binding Buffer 2 and 6 mL of isopropanol were added and mixed. This step releases the nucleic 185 acid bound to the solids into the suspension. The mixture was centrifuged at 3,000 x g for 10 minutes. 186

The supernatant contains the nucleic acid from the solids. The supernatant is then added to the 187 PureYield™ Midi Binding Column for the respective sample and treated independently as described 188

above. 189

A sample of 120 mL of wastewater was centrifuged at 3000 x g for 30 minutes to pellet any particulate 192 material. The supernatant was carefully decanted and then mixed with 12 g of PEG 8000 and 2.7 g of 193 NaCl that were dissolved by gentle mixing. Samples were then centrifuged at 11,400 x g for two hours to 194 pellet viral material. The supernatant was carefully removed via pipetting on the side of the tube opposite 195 . 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)

The copyright holder for this preprint this version posted May 10, 2021. Table S1 lists the primers and probes used in this study, including primer and probe sets for the detection 207 Corp.) to create Positive Control RNA. The linearized plasmid DNA containing the N and E genes, or the 215 in-vitro transcribed RNA were quantified using droplet digital PCR and used as a quantification control. 216

Linear quantification DNA and positive control RNA for PMMoV and MS2 were generated and 217 quantified using the same methods described above. 218 219 220 . 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint

TNA isolated from wastewater was used to perform RT-qPCR using the SARS-CoV-2 RT-qPCR 222 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint

Concordance between trend analysis using the three SARS-CoV-2 targets (N1, N2 and E) was evaluated 245 using Kendall's coefficient of concordance (W). Correlation between weekly normalized SARS-CoV-2 246 levels from wastewater and number of new clinical cases (7 day moving average) was determined using 247

Kendall's tau correlation coefficient. formatted into a BED file. With aligned, primer-trimmed reads in BAM format, the bam-readcount tool 293

(https://github.com/genome/bam-readcount) was used to collect base composition information at the 294 . 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 preprint this version posted May 10, 2021. 

Operating under the assumption that wastewater is likely to contain RT-qPCR inhibitors, we formulated 302 enzyme mixes containing MMLV-RT enzyme and Taq DNA polymerase (with hot-start chemistry) that 303 would be resistant to PCR inhibitors. Next, we designed separate multiplexed RT-qPCR assays to detect 304 the nucleocapsid gene fragment N1 and N2 (as described by the CDC: 305 https://www.cdc.gov/coronavirus/2019-ncov/lab/rt-pcr-panel-primer-probes.html) and the envelope gene We sought to optimize an appropriate amplicon length from the IAC template to provide sufficient 319 sensitivity to detect RT-qPCR inhibition. An in vitro transcribed Luciferase RNA (Promega Corp.) was 320 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint 13 used as a template and four different amplicon lengths (93 bp, 285 bp, 310 bp and 435bp) were tested. 321

Humic acid, a known reverse transcriptase and DNA polymerase inhibitor 43 , was titrated from 0 µg/mL 322 to 125 µg/mL using a two-fold dilution series, and inhibition was assessed by the difference in Ct value 323 with or without humic acid (Ct). The same hydrolysis probe was used in all cases. As expected, we 324 found a correlation between the amplicon length and sensitivity to humic acid (Fig. 1A) 200,0000 copies, as the observed R 2 for all the three targets were ≥ 0.99 (Fig. 1B) . 337

The assay was also tested for specificity with other coronaviruses and respiratory pathogens. The assay 338 was found to be specific for detection of SARS-CoV-2 and PMMoV (Table S2) . 339 . 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. hypothesized that by utilizing a direct capture method to bind total nucleic acid (TNA) to a silica-based 352 affinity resin in place of a method that is selective for intact viral particles, we may be able to eliminate 353 the viral concentration step which is often a cause of technical variability and low recovery. Direct TNA 354 isolation would be unbiased toward intact, partially intact, or free viral RNA. We developed a simple, 355 rapid, highly efficient, and modular alternative to existing wastewater RNA purification methods. The 356 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint primary concentration method utilizes raw sewage, to which chaotropic agents are added to allow binding 357 on to a silica matrix (PureYield™ Midi Binding column) by applying vacuum. The captured nucleic acid 358 is then subjected to successive alcohol washes (to remove RT-qPCR inhibitors that may have co-purified 359 with the nucleic acid) and subsequently eluted in water (Fig. 2) . The eluted nucleic acid can then be 360 further processed in a second step with either a spin column with Wizard ® Enviro Wastewater TNA kit or 361 with an automated nucleic acid purification system for the Maxwell ® RSC Enviro Wastewater TNA kit 362 (Fig. 2) . 

The direct-nucleic acid capture method was first tested for its ability to eliminate RT-qPCR inhibitors. CoV-2 RNA were similar with both purification methods (Fig. 3D) . These data indicate that both manual 389 and automated nucleic acid purification procedures can be used to extract viral genetic material at similar 390 extraction efficiencies. 391 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint 20 mL, 10 mL, and 5 mL) using the column-based manual concentration/purification scheme (Wizard ® ) 407 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint outlined above and performed all final elutions in the same volume (80 µL) of nuclease-free water. The 408 adjusted volumes of Protease solution, binding buffers 1 and 2 and isopropanol used for each starting 409 volume are outlined in Fig. 4A . MS2 bacteriophage was used as a viral spike-in recovery control. 410

The shift in Ct values from the NTC reactions (ΔCt) for the IAC exhibited a sample volume-dependent 411 increase in Ct value indicating a higher concentration of inhibitors may be co-purifying when larger 412 volumes are used. However, the ΔCt was <1 Ct for all the volumes tested, indicating co-purification of 413 RT-PCR inhibitors was not high enough to significantly impact data interpretation (Fig 4B) . We also 414 analyzed the amounts of PMMoV, MS2 and SARS-CoV-2 (N1). As expected, all targets showed a 415 volume-dependent enrichment of genetic material. (Fig 4C-E) . Though we can accurately detect SARS-416

CoV-2 signal from 5 ml of sample during the current sampling period when SARS-CoV-2 clinical cases 417 are high, use of 40 ml sample volume will allow sufficient assay sensitivity when viral loads are lower. 418 419 420 . 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 preprint this version posted May 10, 2021. 

shown are means ± SD (n= 3).

Wastewater, in addition to containing fecal matter and water, is also composed of cellulosic material from 432 toilet and tissue paper that may act as a substrate on which nucleic acids and nucleoprotein complexes can 433 aggregate. This material may form a large part of the suspended solids present in wastewater. SARS-434

CoV-2 may not be present as intact virions in wastewater, but the observation that the genetic material is 435 readily detectable indicates that the viral RNA is likely present in ribonucleoprotein complexes, which 436

shield it from nucleases that are presumably present in wastewater. In addition, detergents and chaotropic 437 agents present in wastewater may also cause structural changes to proteins causing association with 438 suspended solids. Therefore, we reasoned that a proteolytic cleavage step may be able to release some of 439 the viral genetic material associated with solids. 440

We measured viral nucleic acid extraction efficiency using a procedure that included alkaline protease 441 treatment. 40 mL wastewater samples were either treated with alkaline protease or left untreated and 442 processed as described in the Methods section. MS2 phage was also spiked into the samples. Percent 443 recoveries of MS2 nucleic acid were found to be 20% for untreated samples and 40% for samples that 444

were treated with alkaline protease (Fig. 4F) . Similarly, we also observed a two-fold increase in 445 extraction of SARS-CoV-2 viral genome units when samples were treated with protease (Fig. 4G) . These 446 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint results indicate that alkaline protease treatment increases yield of viral TNA extraction, most likely by 447 releasing a portion of the viral genomic material associated with suspended solids. 448 449

In the purification workflow described above, suspended solids are removed via a brief centrifugation 451 following the alkaline protease treatment step. Removing suspended solids prevents clogging of the 452 PureYield™ Midi column. We investigated how much viral matter is associated with the pelleted solids 453 using the procedure described in the methods section. The eluted nucleic acid was analyzed for the 454 quantity of PMMoV and SARS-CoV-2 RNA. We observed that about 11.79% of the total amount of 455

PMMoV and 19.27% of the total amount of SARS-CoV-2 genetic material associated with the solids 456 ( Fig. 5A-B) . If maximal nucleic acid yield is desired, the solids-associated supernatant fraction can be 457 combined with the wastewater supernatant fraction to purify nucleic acid from the total pooled 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. 

Comparison of viral nucleic acid recovery: Direct Capture method vs PEG/NaCl precipitation 466 Using the optimized method described above, we compared the direct capture method with PEG/NaCl 467 precipitation method for the ability to purify SARS-CoV-2 genetic material from wastewater samples. 468 PEG/NaCl is a widely used method for precipitation and concentration of non-enveloped enteric virus 469 such as Poliovirus 28 . 40 mL of wastewater sample was processed for the direct capture method and 120 470 mL of wastewater was processed using PEG/NaCl precipitation. We observed a 20-fold increased 471 extraction efficiency for extracting SARS-CoV-2 RNA (Fig. 6A ) when using the direct capture method 472 compared to the PEG/NaCl protocol. For MS2 (viral spike control) we observed an extraction efficiency 473 of 3.76 ± 1.88% for the PEG/NaCl method and 39.67 ± 10.66% for the direct capture method (Fig 6B) . 474

We also determined the percentage recovery for two human coronaviruses (OC43 and 229E) for the direct 475 capture method and they were 63.13 ± 4.16%, 40.09 ± 10.89% respectively (data not shown). 476

The direct capture purification workflow utilizing 40 ml of wastewater sample and a 40ul elution volume 477 has a concentration factor of 1000. With a LOD of 5 copies for the detection of SARS-CoV-2 by RT-478 qPCR reaction, and 1000-fold concentration in the purification process, the assay sensitivity is around 1 479 viral genome copies/ml. This level of sensitivity will be sufficient for trend analysis using WBE. 480 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint 

The levels of SARS-CoV-2 RNA from three wastewater treatment plants in Dane County, Wisconsin 489 from mid-October, 2020 to early January, 2021 were determined using the Wizard ® Enviro Wastewater 490 TNA kit described above. Samples were processed and analyzed weekly. The levels of RT-qPCR 491 inhibitors present in the TNA samples (as assessed by shift in the Ct value (Ct) of IAC in sample wells 492 compared to NTC wells) were not notable, as Ct values were less than 0.5 for all wastewater samples 493 (Fig. 7A) . 494 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint

We analyzed SARS-CoV-2 RNA levels using N1, N2, and E targets (Fig. 7B ). All samples were positive 495 for all three SARS-CoV-2 targets within the period tested. We calculated the degree of concordance 496 between the three SARS-CoV-2 targets for the three WWTPs over the sampling period using Kendall's 497 coefficient of concordance, W. Kendall's W ranges from 0 (no agreement) to 1 (full agreement). We 498 observed statistically significant concordance between the three SARS-CoV-2 targets for the three 499

WWTPs (W=0.97 for WWTP-1, W=0.96 for WWTP-2 and W=0.78 for WWTP-3). 500

WWTP-1 served a small community of around 10,000 people. We normalized SARS-CoV-2 levels with 501

PMMoV, which has been proposed to account for differential dilution and degradation rates over time. 502

We compared SARS-CoV-2 RNA signals in wastewater to the level of COVID-19 cases (7-day moving 503 average) declared by the municipality and analyzed the correlation between the two, resulting in a 504

Kendall's Tau coefficient of 0.33, with a p-value of 0.08 (Fig 7C) . The peak of SARS-CoV-2 genetic 505 signal observed in the wastewater is concurrent with the peak of positive SARS-CoV-2 reported in mid-506

November of 2020. Even with this limited data set we see the potential for wastewater-based surveillance 507 in assessing community-wide spread of the disease. 508 509 . 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. To determine if our direct capture purification method yielded nucleic acid compatible with next-519 generation sequencing, we prepared sequencing libraries with a subset of wastewater samples using a 520 commercially available SARS-CoV-2 amplicon panel and associated library preparation kit 46,47 . 521

For this proof of concept, we focused on three samples: wastewater from WWTP-2 in November of 2020 522

and January 2021, as well as wastewater collected from WWTP-3 in December of 2020. Because of the 523 proximity of the two collection sites to one another, we reasoned that these samples could be compared 524 broadly for the purpose of identifying SARS-CoV-2 variants, while also demonstrating the robustness of 525 the method to different wastewater treatment regimens from different facilities. 526

We compared total nucleic acid or DNase-treated nucleic acid as input into the library preparation 527 workflow. Most libraries had greater than 1 million reads, but for comparison of depth metrics, the total 528 number of reads was normalized across libraries by randomly subsampling to 650,000 total paired reads. 529

We aligned the subsampled, filtered sequencing reads to the SARS-CoV-2 genome and measured the 530 percentage of filtered reads aligned and the depth of coverage across amplicons tiling the SARS-CoV-2 531 genome ( Fig. 8A-B) . Samples that were DNase-treated had higher percentage of filtered reads aligning to 532 the SARS-CoV-2 genome and greater depth of coverage compared to total nucleic acid libraries. Because 533

wastewater is a heterogeneous sample, it is not surprising that the amplification reaction resulted in off-534 . 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint target amplification. This data shows that DNase treatment of wastewater total nucleic acid improves 535 next-generation sequencing quality from the SARS-CoV-2 amplicon panel tested. 536

We aligned the reads a second time to a combined reference genome containing both SARS-CoV-2 and 537 human reference genomes to examine the proportion of aligned, filtered reads attributable to 538 contamination by human genetic material. Surprisingly, we found that the percentage of filtered reads 539 aligning to the human genome increased after DNase treatment (3.5%-9.4% for total nucleic acid and 540 6.3%-16.5% for DNase-treated samples). However, the proportion of non-human contamination reduced 541 by DNase treatment makes this the preferred protocol. 542

Finally, using all available reads, we analyzed the aligned sequencing reads from the DNase-treated 543 samples to look for the presence of previously reported SARS-CoV-2 variants of concern associated with 544 widespread viral strains. The variants that we detected (Supplemental file 1) were consistent with known 545 variants in Dane County, Wisconsin during the sampling period 36 . Virtually all the SARS-CoV-2 546 sequences contained mutations corresponding to Nextstrain clade 20A. Mutations found in Nextstrain 547 clades 20B, 20C, and 20G were also present, though mutations specific to strain 20B were present at 548 2.0% frequency in January, below the 10% frequency cutoff we had set for making a positive call. 549

Although merely suggestive due to small sample size, this does correspond to Nextstrain reported data 550 which also suggests a decrease in the prevalence of this strain over the sampling period. More 551 transmissible strains were not observed in this data, consistent with Nextstrain reports for such strains 552 initially appearing in this geographic area approximately 2 months after the sampling period. 553 . 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. In this study, we describe a convenient, high throughput, robust, and consistent method to directly 564 capture, concentrate, and detect total nucleic acids (TNA) from wastewater using silica based PureYield ® 565 . 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint columns and optimized RT-qPCR. This method offers ease-of-use and minimizes the need for specialized 566 laboratory equipment. In addition, the method achieves consistent recovery rates and significant reduction 567 in RT-qPCR inhibitors. 568

During the course of this study, alternative direct capture methods for extraction of SARS-CoV-2 RNA 569 from wastewater have been published, highlighting the simplicity of the general workflow 48 . Direct 570 capture methods have also been applied in large-scale interlaboratory method assessment studies where 571 they have shown superior and consistent performance over other methods including PEG/NaCl 572 precipitation, centrifugal ultrafiltration, or charged membranes 49 . By using a protease treatment step, we 573 are able to release a significant portion of viral nucleic acid associated with the solids in the wastewater. 574

In addition, chaotropes and alcohols provide an effective nucleic acid binding environment for capture on 575 a silica matrix. The two-step, modular purification strategy described in this work simplifies the 576 workflow for users processing either small or large amounts of wastewater samples. The flexibility of the 577 method and compatibility of the resulting nucleic acid with downstream analysis by RT-qPCR and 578 SARS-CoV-2 sequencing allow for straightforward adoption for WBE-based viral surveillance 579

Throughout the early course of the COVID-19 pandemic, the ramifications of not having nationwide 581 surveillance systems in place were observed. WBE and clinical diagnostic testing each can provide 582 structured surveillance systems. WBE can complement clinical diagnostic testing by independently 583 confirming prevalence of disease communities and possibly providing an early warning for future viral 584 outbreaks. WBE also provides a low-cost tool to understand community spread in low resource areas. 585 Similar to diagnostic clinical testing, WBE has experimental limitations (uncertainties related to timing 586 and quantities of viral and viral nucleic acid shedding, RNA stability, effect of temperature, and sample 587 processing techniques) that need to be well understood before using acquired data to inform 588 epidemiological and public health efforts around the globe. 589 . 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint

As the COVID-19 pandemic has progressed, SARS-CoV-2 genetic variants have arisen, often leading to 590 increased transmissibility, concern about immune evasion, and subsequent outbreaks 50 . With the 591 emergence of new strains, it will be informative to see how this rapidly evolving method is able to help 592 understand the spread of variants within communities and guide health authorities to take appropriate 593 measures 51-53 . 594 595

Sequencing data are available upon request of the corresponding author. 597 . 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 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)

The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint

Resources, Writing -review & editing. Sameer Moorji

Supervision, Writing -review & editing

A Novel Coronavirus from Patients with Pneumonia in China

The molecular virology of coronaviruses

Origin and evolution of pathogenic coronaviruses

A pneumonia outbreak associated with a new coronavirus of probable bat origin

Temporal dynamics in viral shedding and transmissibility of COVID-19

SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of 632 viral shedding, and infectiousness: a systematic review and meta-analysis. The Lancet Microbe

Evidence Supporting Transmission of Severe Acute 635

Respiratory Syndrome Coronavirus 2 While Presymptomatic or Asymptomatic. Emerging 636 Infectious Disease journal 26

Early viral clearance and antibody kinetics of COVID-19 among asymptomatic 638 carriers

Detection of SARS-CoV-2 in Fecal Samples From Patients With Asymptomatic 640 and Mild COVID-19 in Korea

Covid-19: identifying and isolating asymptomatic people helped eliminate virus in 643

Detection of SARS-CoV-2 in Different Types of Clinical Specimens

The presence of SARS-CoV-2 RNA in the feces of COVID-19 patients

Viral load dynamics and disease severity in patients infected with SARS-CoV-2

retrospective cohort study

Virological assessment of hospitalized patients with COVID-2019

Viral RNA Load in Mildly Symptomatic and Asymptomatic Children with 654

Emerging Infectious Disease journal 26

Interpreting Diagnostic Tests for SARS-CoV-2

Wastewater-based epidemiology: current status and future prospects

Future perspectives of wastewater-based epidemiology

Monitoring infectious disease spread and resistance to the community level

Detection of pathogenic viruses in sewage provided early warnings of hepatitis 665 A virus and norovirus outbreaks

Role of environmental poliovirus surveillance in global polio eradication and 668 beyond

Environmental surveillance for polioviruses in the Global Polio Eradication 670

Coronavirus-2 RNA in Sewage and Correlation with Reported COVID-19 Prevalence in the Early 673

Stage of the Epidemic in The Netherlands

First detection of SARS-CoV-2 in untreated wastewaters in Italy

First confirmed detection of SARS-CoV-2 in untreated wastewater in Australia: 678 A proof of concept for the wastewater surveillance of COVID-19 in the community

Measurement of SARS-CoV-2 RNA in wastewater tracks community infection 681 dynamics

SARS-CoV-2 RNA in wastewater anticipated COVID-19 occurrence in a 683 low prevalence area

Evaluation of lockdown impact on SARS-CoV-2 dynamics through viral 685 genome quantification in Paris wastewaters

International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity

Recent trends on methods for the concentration of viruses from 688 water samples

Comparison of virus concentration methods for the RT-qPCR-based recovery of 691 murine hepatitis virus, a surrogate for SARS-CoV-2 from untreated wastewater

SARS-CoV-2 in wastewater: State of the knowledge and research needs

Quantitative PCR for determining the infectivity of 697

UV-B radiation, and singlet oxygen: advantages 698 and limitations of an enzymatic treatment to reduce false-positive results

fastp: an ultra-fast all-in-one FASTQ preprocessor

Initial sequencing and analysis of the human genome

The Human Genome Browser at UCSC

The Genome Analysis Toolkit: A MapReduce framework for analyzing next-707 generation DNA sequencing data

Nextstrain: real-time tracking of pathogen evolution

The UCSC Table Browser data retrieval tool

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) The copyright holder for this preprint this version posted

Pepper Mild Mottle Virus, a Plant Virus Associated with Specific Immune 715

Abdominal Pains, and Pruritus in Humans

Pepper mild mottle virus as a water quality indicator

Cross-assembly phage and pepper mild mottle virus as viral water quality 720 monitoring tools-potential, research gaps, and way forward

Mottle Virus</em> as an Indicator of Fecal Pollution

PCR inhibitors -occurrence, properties 726 and removal

Persistence of SARS-CoV-2 in Water and Wastewater

Quantitative microbial risk assessment (QMRA) of occupational 730 exposure to SARS-CoV-2 in wastewater treatment plants. Science of The Total Environment

A rapid assessment of wastewater for genomic surveillance of SARS-CoV-2 733 variants at sewershed scale in Louisville, KY. 2021

High-throughput sequencing of SARS-CoV-2 in wastewater provides 736 insights into circulating variants

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) The copyright holder for this preprint this version posted May 10

Reproducibility and sensitivity of 36 methods to quantify the SARS-CoV-2 742 genetic signal in raw wastewater: findings from an interlaboratory methods evaluation in the

Water Research & Technology

SARS-CoV-2 Viral Variants-Tackling a Moving 746

Genome Sequencing of Sewage Detects Regionally Prevalent SARS-748

CoV-2 Variants

Monitoring SARS-CoV-2 Circulation and Diversity through Community 750

Wastewater Sequencing, the Netherlands and Belgium. Emerging Infectious Disease journal

Rapid screening for SARS-CoV-2 variants of concern in clinical and 753 environmental samples using nested RT-PCR assays targeting key mutations of the spike protein

Amplicon library preparation guidance, basic analysis guidance, and MiniSeq sequencing were provided 599 by Swift Biosciences, Inc (https://swiftbiosci.com). We thank Paul Muschler for providing valuable 600 feedback. We acknowledge the staff of Madison Metropolitan Sewage District (MMSD), Village of 601Oregon, WI and City of Sun Prairie, WI for their support and providing wastewater samples. We also 602 thank the various research groups, public health, and commercial labs with whom we have had many 603 productive discussions on this topic in these difficult times. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 10, 2021. 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 preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.06.21256753 doi: medRxiv preprint