key: cord-0835557-z29a81p7
authors: Ransome, E.; Hobbs, F.; Jones, S.; Coleman, C.; Harris, D.; Woodward, G.; Bell, T.; Trew, J.; Kolarevic, S.; Kracun-Kolarevic, M.; Savolainen, V.
title: No evidence for environmental transmission risk of SARS-CoV-2 in the UK's largest urban river system: London as a case study
date: 2022-03-16
journal: nan
DOI: 10.1101/2022.03.16.22272465
sha: ab2fee4dbb5ec28cb3c58328b9fea9d5aa2b2035
doc_id: 835557
cord_uid: z29a81p7

The presence of SARS-CoV-2 in untreated sewage has been confirmed in many countries but its incidence and infection risk in contaminated freshwaters is still poorly understood. The River Thames in the UK receives untreated sewage from 57 Combined Sewer Overflows (CSOs), with many discharging dozens of times per year. We investigated if such discharges provide a pathway for environmental transmission of SARS-CoV-2. Samples of wastewater, surface water, and sediment collected close to six CSOs on the River Thames were assayed over 8 months for SARS-CoV-2 RNA and infectious virus. Bivalves were sampled as sentinel species of viral bioaccumulation. Sediment and water samples from the Danube and Sava rivers in Serbia, where raw sewage is also discharged in high volumes, were assayed as a positive control. We found no evidence of SARS-CoV-2 RNA or infectious virus in UK samples, in contrast to RNA positive water and sediment samples from Serbia. Furthermore, we show that infectious SARS-CoV-2 inoculum is stable in Thames water and sediment for < 3 days, while RNA remained detectable for at least seven days. This indicates that dilution of wastewater likely limits environmental transmission, and that infectivity should be embedded in future risk assessments of pathogen spillover.

Early detection and containment of the SARS-CoV-2 virus is essential to contain 36 community outbreaks of COVID-19 (1). While the primary route of viral transmission 37 between humans is via exposure to respiratory fluids carrying infectious virus (2,3), 38 evidence of faecal-oral transmission has raised concerns regarding possible 39 environmental transmission to humans and wildlife through spillover from sewage 40 (4-6). Numerous studies have recorded SARS-CoV-2 in faeces at up to 10 7 genome 41 copies/ml (reviewed by (7)). Faulty sewerage systems have previously been linked to 42 earlier SARS-CoV-1 (another highly pathogenic human coronavirus) outbreaks (8), 43 and the presence and infectious potential of other coronaviruses in water and 44 sewage ranges from days to weeks (9). Together, these studies indicate 45 transmission of SARS-CoV-2 via sewage is a potential ongoing and future concern 46 for SARS-CoV-2 outbreaks (10). 47 48 Like other coronaviruses, SARS-CoV-2 RNA has been detected in wastewater in 49 multiple countries (reviewed by (5)), and genome concentrations have correlated 50 . 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 March 16, 2022. ; https://doi.org/10.1101/2022.03.16.22272465 doi: medRxiv preprint positively with the number of human cases within the catchment (11). This indicates 51 that wastewater-based epidemiology could be an efficient way to monitor SARS-52

CoV-2 dynamics in human populations at large scales. SARS-CoV-2 RNA has also 53 been reported in rivers, due to inadequate wastewater treatment or sewage spillover 54 prior to treatment (12-14), suggesting a potential route for transmission of SARS-55

CoV-2 to humans and wildlife, particularly in urban areas. 56 57

The River Thames is the UK's second longest river, with a catchment covering over 58 16 000 km 2 (15). Its Greater London area houses about 14 million people (16) -one 59 fifth of the entire UK population -with many more visiting the area daily, and it 60 provides about 2/3 of London's water supplies (17). The Thames supports many 61 species of wildlife and is also used for recreation, which brings humans and 62 potential hosts and animal vectors of disease into close contact. It also acts as the 63 outlet for 57 Combined Sewer Overflows (CSOs; (18)), which release raw 64 (untreated) and processed sewage, directly into the river. These overflows are 65 designed to reduce the risk of sewage flooding homes and businesses and, although 66 they operate throughout the year, they are used particularly during periods of heavy 67 rain in the winter (15), when SARS-CoV-2 is also at its seasonal peak in the human 68 population (20). Although recent improvements to the Thames sewerage network 69 have reduced sewage discharges from around 40 million tonnes in 2011 to 18 million 70 tonnes per year (15), individual sewage works are still discharging 3.5 billion litres of 71 untreated sewage a year, with occasions during the initial pandemic in 2020 of more 72 than 1 billion litres being released in one day (21). 73 74 . 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. (which was not certified by peer review) from two different sites (see supplementary Table 1 ). Samples were collected using 152 the methodology described above for the Thames, from sites downstream of CSOs. were used for propagation and infectivity assays of MHV using the manufacturers 164 protocol using DMEM high glucose with 10% horse serum. All viral stocks were 165 . 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. As this method has only been validated for non-enveloped viruses (34), via RNA 177 detection, to validate it for assessing the presence and infectivity of SARS-CoV-2, 10 178 l of Milli-Q water was spiked with murine hepatitis virus (MHV-A59) to a final 179 concentration of 1.5 x 10 4 gc/L, and run through the TFUF to test processing 180 efficiency. Two millilitres of sample was collected at the end of the process, filtered 181 (0.22 µM) and 1.5 ml of sample was used to carry out a TCID 50 infectivity assay, as 182 above. TCID 50 /ml was converted to the expected concentration, and compared to 183 values recovered in spiked water. Replicate runs (n=3) showed the concentration of 184 MHV after PEG precipitation to range from a maintenance of the original 185 concentration in spiked water to 160 x more concentrated than spiked water 186 (average of 55.6 x SE +/-52.2; see supplementary Table 2) . 187

Processing of sediment and bivalve samples 189 . 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 March 16, 2022. ; https://doi.org/10.1101/2022.03.16.22272465 doi: medRxiv preprint Sediment samples were processed using beef extract elution as described by Farkas 190 et al. (35) . Five grams of sediment were added to 15 ml of 3 % beef extract, 2 M 191 sodium nitrate, pH 5.5. Solid matter was removed by centrifugation at 3 000 rpm for 192 10 min and a PEG precipitation was carried out as above (see concentration of viral 193 particles in water). 194

The digestive tissue of bivalves was extracted and processed as described 196

previously (27). In brief, up to 2 g of homogenised digestive tissue was incubated in 197 1 ml 0.2 mg/ml proteinase K solution for 60 min at 37 o C followed by 15 min at 60 o C. 198

The liquid phase was separated by centrifugation at 3 000 rpm for 10 min. Coralville, Iowa, USA) (36) and the limit of detection for SARS-CoV-2 was 216 determined using a curve-fitting method as described by Klymus while RNA was relatively stable over a week-long incubation, our ability to recover 293 infectious virus from the samples (TCID 50 ; Figure 3 ) declined rapidly over the first 3 294 days with no viable virus present after one week. SARS-CoV-2 RNA recovered from 295 sediment samples was lower than that found in water, but it too remained relatively 296 stable over 7 days (RNA; Figure 3 ). In contrast, no viable virus was recovered from 297 . 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. Across 218 samples of CSO water, surface water, river sediment and bivalves 311 collected from the Thames Basin we found no SARS-CoV-2 RNA, despite our study 312 being carried out during two periods of the highest reported cases in London (e.g., 313 11 536 new cases reported on January 11 th and 7 641 on July 15 th , 2021) 314 . 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 March 16, 2022. ; https://doi.org/10.1101/2022.03.16.22272465 doi: medRxiv preprint (https://coronavirus.data.gov.uk/details/cases), and in a region with one of the 315 highest country-wide SARS-CoV-2 cases throughout the pandemic 316 (https://coronavirus.data.gov.uk/details/cases). We found evidence of the human gut 317 bacteriophage, crAssphage, at five out of our 6 sites on the Thames, confirming the 318 presence of sewage in our samples (38), and evidence for the detection of SARS-319

CoV-2 RNA in samples from the Danube and Sava rivers in Serbia, with the same 320 methodology, suggesting that if SARS-CoV-2 was present in the Thames, above our 321 limit of detection, we would have found it. From Serbia, concentrations of the N1 322 gene from water samples collected from the Sava River in the declining phase of the 323 fourth COVID-19 wave, were of the same order of magnitude as those found in 324

December 2020: 5.97 × 10 3 to 1.32 × 10 4 copies/L (13), confirming that our method 325 concentrates SARS-CoV-2 RNA with equal efficiency to other methods. Further, 10x 326 more water was concentrated per sample from the Thames (10 litres) than from 327

Serbia (1 litre). This all suggests that while London sewage works do pump raw 328 sewage into the Thames on a regular basis, the dilution of sewage with storm and 329 surface water, and the tidal nature of the Thames at our sampling sites, unlike the 330 non-tidal Sava and Danube rivers, is sufficient to dilute SARS-CoV-2 RNA to below 331 our limit of detection and reduce the threat of SARS-CoV-2 environmental spillover. 332

In contrast, previous studies that have found SARS-CoV-2 RNA in river water 334 (13,14) are unsurprising as unprocessed sewage is pumped directly into rivers in 335

Ecuador and Serbia, with little or no attempt to neutralise associated microbial 336 communities. These studies have focused on water sampling, in line with standard 337 sampling of surface waters for environmental monitoring (e.g. for bathing water 338 quality (24). Here we provide the first evidence for the accumulation of high 339 . 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. such as MHV, are useful for assessing method performance during monitoring 361 campaigns (49) . MHV belongs to the same genus as SARS-CoV-2, is structurally 362 and morphologically similar (50), and has been shown to have similar decay rates in 363 wastewater (47,51,52). The TFUF-PEG concentration method tested here retains 364 . 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 March 16, 2022. ; https://doi.org/10.1101/2022.03.16.22272465 doi: medRxiv preprint 1% of infectious virus, which equates to up to 160x the concentration of virus 365 present in the initial spiked water. This is within the limits of the 1-25% of infectious 366 virus (as plaque forming units; reviewed by (53)) recovered by other methods. This 367

suggests that while other methods may be superior for more concentrated samples 368 (e.g. sewage influent), TFUF is a valid method for concentration of enveloped RNA 369 viruses from highly diluted samples, such as river water. However, our findings also 370 highlight the need to develop more robust methods for monitoring more fragile, and 371 potentially dangerous enveloped viruses in the environment. 372

While clinical studies of faecal material from hospitalised patients have isolated 374 virulent virus (54), our findings add to recent evidence that detected SARS-CoV-2 375 RNA from the natural environment does not occur as infectious viral particles, and 376 thus do not represent a health hazard (12,55,56). Our laboratory study also provides 377 evidence of relatively rapid degradation of SARS-CoV-2 infectivity in CSO water, 378 while RNA concentrations remain stable. This highlights that infectivity data should 379 be embedded within risk assessments of pathogen spillover, alongside the 380 occurrence data that is amassed by q(RT)-PCR. For SARS-CoV-2, it suggests that 381 while some viral particles may remain infectious long enough to reach surface 382 waters, they are unlikely to accumulate over time. However, further work to confirm 383 that SARS-CoV-2 does not survive in these systems during the colder winter 384 months, when coronaviruses are thought to survive longer (9), is needed. 385

The COVID-19 pandemic has highlighted the dramatic consequences of novel 387

outbreaks of viral pathogens. Public health organizations such as the Centre for 388

Disease Control and Prevention, and the World Health Organisation, have therefore 389 . 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. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 16, 2022. ; https://doi.org/10.1101/2022.03.16.22272465 doi: medRxiv preprint

Evaluation of Two Triplex One-Step qRT-PCR Assays for the Quantification of 537

Human Enteric Viruses in Environmental Samples

Analytical and clinical comparison of Viasure (CerTest Biotec) 541 and 2019-nCoV CDC (IDT) RT-qPCR kits for SARS-CoV2 diagnosis

Reporting the limits of detection and quantification for environmental DNA 545 assays. Environmental DNA

CrAssphage as a Potential 547

Human Sewage Marker for Microbial Source Tracking in Southeast Asia

Environmental Science and Technology Letters

Co-550 quantification of crAssphage increases confidence in wastewater-based 551 epidemiology for SARS-CoV-2 in low prevalence areas

CoV-2 RNA in wastewater anticipated COVID-19 occurrence in a low 555 prevalence area

Serological assays for severe acute respiratory 558 syndrome coronavirus 2 (SARS-CoV-2)

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

First 585 confirmed detection of SARS-CoV-2 in untreated wastewater in Australia: A 586 proof of concept for the wastewater surveillance of COVID-19 in the 587 community

The species Severe acute respiratory syndrome-related coronavirus: 590 classifying 2019-nCoV and naming it SARS-CoV-2

Environmental and decontamination issues for human 593 coronaviruses and their potential surrogates

Genomic Modeling as an 596 Approach to Identify Surrogates for Use in Experimental Validation of SARS

CoV-2 and HuNoV Inactivation by UV-C Treatment. Frontiers in Microbiology

Current 602 Opinion in Environmental Science and Health

Infectious SARS-CoV-2 in 604

Feces of Patient with Severe COVID-19. Emerging Infectious Diseases

Detection of SARS-CoV-2 in raw and treated

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 March 16

Persistence of SARS-CoV-2 in Water and Wastewater. Environmental Science 612 and Technology Letters

Detection of SARS-CoV-2

RNA in bivalve mollusks and marine sediments

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

prioritized scientific research to enhance our ability to rapidly identify, track and 390 contain novel human pathogens. Although we know the presence of SARS-CoV-2 391 RNA in raw sewage is abundant enough to be used to monitor levels of infection in 392 human populations, our study indicates that the quantities of raw and processed 393 sewage reaching the Thames, is low enough to reduce the threat of environmental 394 spillover of SARS-CoV-2 from faecal sources. Reduced threat does not mean, 395 however, that the Thames is safe. It is still unclear how many viral particles are 396 needed to cause an infection in humans, and although we could not detect infectious 397 SARS-CoV-2 in the Thames, those particles might still be there, below our limit of 398 detection. They may also be present in non-tidal areas, where river water is not 399 diluted daily. What is needed is resilient and modernised sewerages to keep our 400 rivers uncontaminated. High concentrations of SARS-CoV-2 RNA in the Danube and 401Sava rivers in Serbia is concerning, especially as SARS-CoV-2 RNA has now been 402 found in mollusc tissue (e.g. (57)). While our study focuses on SARS-CoV-2, rivers 403 are potentially conduits to multiple pathogens and disease transmission via raw 404 sewage spillover will remain a threat as long as water companies continue to 405 Secondary transmission of SARS-CoV-2 through wastewater: Concerns and 438 . 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. 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. Technology. Royal Society of Chemistry; 2020. p. 1213-6. 584 . 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. (which was not certified by peer review)