key: cord-0981611-iywsf5wt authors: Renninger, N.; Nastasi, N.; Bope, A.; Cochran, S. J.; Haines, S. R.; Balasubrahmaniam, N.; Stuart, K.; Bivins, A.; Bibby, K.; Hull, N. M.; Dannemiller, K. C. title: Indoor dust as a matrix for surveillance of COVID-19 outbreaks date: 2021-01-11 journal: nan DOI: 10.1101/2021.01.06.21249342 sha: 21f93f5ebeec8c815b8a0b1855cc799e2dacffaa doc_id: 981611 cord_uid: iywsf5wt Ongoing disease surveillance is a critical tool to mitigate viral outbreaks, especially during a pandemic. Environmental monitoring has significant promise even following widespread vaccination among high-risk populations. The goal of this work is to demonstrate molecular SARS-CoV-2 monitoring in bulk floor dust and related samples as a proof-of-concept of a non-invasive environmental surveillance methodology for COVID-19 and potentially other viral diseases. Surface swab, passive sampler, and bulk floor dust samples were collected from rooms of individuals infected with COVID-19, and SARS-CoV-2 was measured with quantitative reverse transcription polymerase chain reaction (RT-qPCR) and two digital PCR (dPCR) methods. Bulk dust samples had geometric mean concentration of 159 copies/mg-dust and ranged from non-detects to 23,049 copies/mg-dust detected using ddPCR. An average of 88% of bulk dust samples were positive for the virus among detection methods compared to 55% of surface swabs and fewer on the passive sampler (19% carpet, 29% polystyrene). In bulk dust, SARS-CoV-2 was detected in 76%, 93%, and 97% of samples measured by qPCR, chip-based dPCR, and droplet dPCR respectively. Detectable viral RNA in the bulk vacuum bags did not measurably decay over 4 weeks, despite the application of a disinfectant before room cleaning. Future monitoring efforts should further evaluate RNA persistence and heterogeneity in dust. This study did not measure virus viability in dust or potential transmission associated with dust. Overall, this work demonstrates that bulk floor dust is a potentially useful matrix for long-term monitoring of viral disease outbreaks in high-risk populations and buildings. . CC-BY-NC-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 January 11, 2021. rooms of individuals infected with COVID-19, and SARS-CoV-2 was measured with 48 quantitative reverse transcription polymerase chain reaction (RT-qPCR) and two digital PCR 49 (dPCR) methods. Bulk dust samples had geometric mean concentration of 159 copies/mg-dust 50 and ranged from non-detects to 23,049 copies/mg-dust detected using ddPCR. An average of 51 88% of bulk dust samples were positive for the virus among detection methods compared to 55% 52 of surface swabs and fewer on the passive sampler (19% carpet, 29% polystyrene). In bulk dust, 53 SARS-CoV-2 was detected in 76%, 93%, and 97% of samples measured by qPCR, chip-based 54 dPCR, and droplet dPCR respectively. Detectable viral RNA in the bulk vacuum bags did not 55 measurably decay over 4 weeks, despite the application of a disinfectant before room cleaning. 56 Future monitoring efforts should further evaluate RNA persistence and heterogeneity in dust. 57 This study did not measure virus viability in dust or potential transmission associated with dust. 58 Overall, this work demonstrates that bulk floor dust is a potentially useful matrix for long-term 59 monitoring of viral disease outbreaks in high-risk populations and buildings. CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint Introduction 75 The spread of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) 76 reached pandemic designation in March 2020 and has since resulted in more than 75 million 77 cases of COVID-19 and 1.6 million deaths documented worldwide as of December 21, 2020 (1). 78 Both symptomatic and asymptomatic carriers shed the virus into the environment (2-4). Viral 79 particles are emitted primarily via respiratory droplets and aerosols, and persist on surfaces 80 indoors (4-6). SARS-CoV-2 viability has been characterized after deposition onto several 81 surface types (6). In one study, viable virus was detected on plastics and stainless steel up to 72 82 hours after application (5). Other studies have demonstrated respiratory viruses can contaminate 83 environmental dust near infected individuals (7-9). These viral shedding routes together with 84 persistence indoors and in environmental dust implicate potential viral contamination of indoor 85 dust near infected individuals (10). . CC-BY-NC-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 January 11, 2021. There is a critical need for targeted, efficient, and inexpensive methods to monitor SARS- CoV-2 and other viruses long-term to identify potential viral outbreaks prior to extensive spread. Fecal shedding of SARS-CoV-2 provides the basis for large-scale viral monitoring in wastewater 89 systems (11) (12) (13) (14) . However, more targeted monitoring efforts are needed for indoor 90 environments, especially those housing vulnerable populations such as congregate care facilities. 91 We propose that the detection of SARS-CoV-2 RNA in indoor dust can be used for continued 92 environmental surveillance of novel coronavirus, SARS-CoV-2. Targeted monitoring of dust in 93 high-concern buildings could complement broader population-level monitoring approaches. This 94 strategy could then be extended to other viruses of concern. Our goal is to demonstrate that 95 indoor dust can be used as a matrix for viral outbreak surveillance. We measured SARS-CoV-2 using quantitative reverse transcription polymerase chain 99 reaction (RT-qPCR), chip-based digital PCR (dPCR), and droplet digital PCR (ddPCR) in 100 samples of bulk dust, passive surface samples, and surface swabs from rooms of individuals with 101 COVID-19. In bulk dust, the SARS-CoV-2 viral concentration had a geometric mean value of 102 159 copies/mg-dust and ranged from non-detects to 23,049 copies/mg-dust ( Figure 1A ). We 103 detected SARS-CoV-2 RNA in 89% of bulk dust, 55% of surface swabs, and 38% of passive 104 surface sampler samples (average among all three detection methods used). The ddPCR method 105 detected viral RNA in 97% of bulk dust samples compared to 93% for the chip-based dPCR and 106 76% for RT-qPCR ( Figure 1B ). The COVID-19 isolation rooms were treated with a chlorine-based disinfectant prior to 108 dust collection as part of the normal cleaning process, and the disinfectant is expected to largely 109 . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint inactivate the virus through reactions with the viral capsid (15). The bags were stored in the 110 laboratory at room temperature after collection. Triplicate subsamples were extracted and viral 111 RNA measured immediately upon collection and once per week for 4 weeks. Viral RNA did not 112 measurably decay over 4 weeks in the vacuum bags (regression R 2 =0.009, p=0.47) (Figure 2A ). The coefficient of variance (CoV) for copies/mg-dust ranged from 73.5-313.4% within each CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint 6 fractions within a vacuum bag. Therefore, multiple samples should be taken from a bag to more 133 rigorously quantify the viral genetic signal or homogenization methods should be developed that 134 comply with biosafety standards. Additionally, RNA and dust persistence in the environment 135 should be considered when determining if the outbreak occurred recently or in the past. Limitations of this study include that we did not measure the viability of SARS-CoV-2 in 150 the dust samples due to biosafety constraints, although this is not needed for surveillance. Also, 151 our small sample size from rooms occupied by infected students may not be representative of 152 other buildings and occupancy conditions, and samples were collected after a known infection as 153 opposed to before. More information is needed on how representative each dust sample would 154 be for a specific population and different occupancy levels. We were unable to sieve or 155 . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint otherwise homogenize the dust due to biosafety concerns, which likely resulted in variability CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint 13 ddPCR 293 Droplet digital PCR was performed using the Bio-Rad QX200 system along with a 294 C1000 Touch Thermal Cycler (Biorad, Hercules, CA). SARS-CoV-2 RNA was detected and 295 quantified using the N1 assay previously described. Inhibition was assessed by spiking a subset . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint The 95% limit of detection for the N1 assay was determined to be 3.3 gene copies per 316 ddPCR reaction using a ten-replicate dilution series of synthetic SARS-CoV-2 RNA control 317 material (MT188340, Twist Bioscience, San Francisco, CA) with a cumulative Gaussian 318 distribution fit to the observed proportion of the replicates positive along the dilution gradient. There was no evidence of inhibition as no difference was observed in the quantification of BRSV 320 RNA in sample extracts compared to the BRSV positive controls (two-tailed t-test, p=0.19). Statistical and data analysis 323 Our goal was to compare measurement of SARS-CoV-2 in bulk dust, on surface swabs, 324 and on a passive sampler using three different measurement methods. Each vacuum bag of dust 325 was sampled and extracted in triplicate at each time point (immediately after collection and 1,2,3 326 and 4 weeks post collection). All three detection methods (qPCR, dPCR and ddPCR) analyzed 327 the same sample extractions for all sample types. Detection limit information is described above 328 for each detection method. The geometric mean was reported for quantification of SARS-CoV-2 329 RNA present in samples using each method due to the logarithmic nature of PCR-based data. Potential RNA decay over the 4 week time period was evaluated in bulk dust with a regression 331 analysis on the ddPCR data transformed with the natural logarithm. The dataset is available at 332 https://doi.org/10.5061/dryad.3n5tb2rg1. Acknowledgements 335 We are grateful to the isolation room coordinator/manager and staff, as well as the occupants for 336 donation of the samples. We would also like to thank the carpet manufacturer for donation of 337 . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint Tracking COVID-19 with wastewater First confirmed detection of SARS-CoV-2 in untreated 385 wastewater in Australia: A proof of concept for the wastewater surveillance of COVID-19 386 in the community Wastewater-Based Epidemiology: Global 397 Collaborative to Maximize Contributions in the Fight Against COVID-19 COVID-19 surveillance in Southeastern Virginia using 401 wastewater-based epidemiology International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Differences in Viral Disinfection Mechanisms as Revealed by Quantitative Transfection of Echovirus 11 Genomes Aspergillus surveillance project at a large tertiary-care hospital Eight-year surveillance of environmental fungal contamination in hospital operating rooms 411 and haematological units The Effect of Legionella Pneumophila Contamination in 414 the Surface Dust of the Air Ducts of Central Air Conditioning Systemson Indoor Air 415 Computational analysis of SARS-CoV-2/COVID-19 417 surveillance by wastewater-based epidemiology locally and globally: Feasibility, economy, 418 opportunities and challenges Fecal viral shedding in COVID-19 patients: Clinical significance, viral load dynamics and survival analysis well-characterized carpet samples for the passive sampler. 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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint 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 January 11, 2021. . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint . CC-BY-NC-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 January 11, 2021. ; https://doi.org/10.1101/2021.01.06.21249342 doi: medRxiv preprint