key: cord-1036838-rwsu3nln
authors: Barrios, Melina Elizabeth; Díaz, Sofía Micaela; Torres, Carolina; Costamagna, Damián Matías; Fernández, María Dolores Blanco; Mbayed, Viviana Andrea
title: Dynamics of SARS-CoV-2 in wastewater in three districts of the Buenos Aires metropolitan region, Argentina, throughout nine months of surveillance: A pilot study
date: 2021-08-12
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2021.149578
sha: 65431f23762a54dc3dde20185cd56f1d000c6fb7
doc_id: 1036838
cord_uid: rwsu3nln

In the current pandemic of COVID-19, sewage surveillance of SARS-CoV-2 genome has been used to complement viral epidemiology in different countries. The aim of this work was to introduce and evaluate this wastewater-based tool in the metropolitan region of Buenos Aires, Argentina. As a pilot study, surveillance of SARS-CoV-2 in wastewater from three districts of this area was performed for more than nine months from June 2020 to April 2021. Viruses present in the samples were concentrated using polyethylene glycol precipitation and quantified using RT-qPCR CDC N1 assay. Virus recovery for SARS-CoV-2 and a potential surrogate, bovine coronavirus Mebus strain, that shares the Betacoronavirus genus and structural characteristics with SARS-CoV-2, were evaluated after concentration and detection procedures. Recovery of both viruses did not differ significantly, with a median for SARS-CoV-2 and BCoV of 0.085 (95% CI: 0.021-0.179) and 0.262 (95% CI: 1.18 × 10-5-0.564) respectively. The concentration of SARS-CoV-2 genome in wastewater ranged from 10 -1 to 10 3 cg/ml, depending on the wastewater treatment plant, type of collection site, viral recovery of the concentration method and the epidemiological situation of the outbreaks. Significant correlations were observed between SARS-CoV-2 concentration in wastewater and reported clinical cases, reinforcing the utility of this approach to monitor the epidemiological status of populations.

Infection of a population caused by microbial agents that are shed via stool and urine can be tracked by wastewater analysis. In the current pandemic of COVID-19, sewage surveillance of SARS-CoV-2 genome has been used to complement viral epidemiology in different countries, which was reviewed recently (Medema et al., 2020a) . Viral detection in wastewater is possible due to the prolonged viral shedding of SARS-CoV-2 in stools of patients even after respiratory samples became negative (Parasa et al., 2020; van Doorn et al., 2020; Wölfel et al., 2020) . The prevalence of SARS-CoV-2 in feces of patients ranges between 15.3 to 83.3% according to different studies and metanalyses (Cheung et al., 2020; Foladori et al., 2020) . Respiratory fluids excreted to the sewer system also might contribute to the viral genome detection in these samples. Wastewater-based surveillance has been applied mostly to non-enveloped viruses (Chen et al., 2020; WHO, 2003; Wyn-Jones et al., 2011) and so it is necessary to adapt concentration methods for enveloped viruses such as SARS-CoV-2. Also, the performance of detection procedures designed for clinical samples must be evaluated when applied to environmental samples.

Epidemiological data based on nasopharyngeal sample tests only gives information about a fraction of the infected persons, either due to the limited testing capacity or the unrecognized asymptomatic infections. However, the analysis of the wastewater could render information about the whole population. Depending on the point of the sewer system selected for locations with limited clinical surveillance, and research purposes (Global Infectious Hazard Preparedness WHO Headquarters (HQ) WHO Worldwide, 2020).

In Argentina, the first imported case of SARS-CoV-2 was detected on March 3 rd , 2020. The main affected area in the country was the Buenos Aires Metropolitan Region (referred to as BAMR), where the communitarian circulation of the virus began on March 23 rd . A lockdown was implemented on March 20 th , which delayed the exponential increase in the number of cases. In September, the BAMR reached a temporary maximum of reported cases and started a stabilization trend while the virus increased its spreading through all the country. By March of 2021, the second wave of the pandemic was beginning, and during April 11-15% of the population had received at least one dose of the vaccination scheme.

The aim of this work was to conduct a pilot wastewater-based epidemiological study of SARS-CoV-2 of three districts of the BAMR between June 2020 to April 2021. The evaluation of the possible correlation between SARS-CoV-2 RNA levels in wastewater and COVID-19 cases could show how useful this tool can be to assist public health responses.

The Buenos Aires Metropolitan Region (BAMR) is a wide territory with a surface of 13,285 km 2 that includes Buenos Aires city and 40 districts that represent approximately 14.8 million inhabitants. The districts of the Region possess different population densities and different kinds of sewer systems. A pilot study to evaluate the usefulness of the wastewater-based epidemiological tool in our region was planned. Raw sewage was weekly collected for more than nine months (from mid-June 2020 to early April 2021) from four sewer systems.

AK and SV represented very crowded communities, close connected, placed in the South of the BAMR both belonging to a superior administrative-territorial division, called Partido San Vicente, from here on, named district San Vicente (Figure 1 ). These sewer networks were selected because they serve well-defined geographic regions and are not impacted by storm discharges. The population registered according to the last census in 2010, was estimated as 35,407 inhabitants in AK location, 21,411 inhabitants in SV location and 59,478 inhabitants in the whole district San Vicente (Instituto Nacional de Estadística y Censos (INDEC), 2012). A new population survey that was meant to be conducted in 2020 was postponed due to the pandemic. Wastewater samples were taken from the inlet pipe of the WWTPs of AK and SV, which serve population equivalents of 20,000 and 15,000, respectively. Then, samples collected at this point of the sewer networks, represent the excretion from 56.47 % of the population of AK, 70.06 % of the population of SV and 58.85 % of the whole district San Vicente.

The other two sampling points in the north of the BAMR, SP and SF, belong to other two different districts and represented sewer systems that serve smaller populations of lowerincome neighborhoods (3,500 and 5,000 population equivalents, respectively). SP samples were taken from the inlet pipe of the WWTP while the samples from SF were collected from a manhole of the sewer system. Grab samples of 250 ml were collected weekly, at the same hour in the morning, in these four locations. Samples were refrigerated at 10°C and transported to the laboratory, to be immediately processed.

Samples were pasteurized in a 60°C water bath for 90 min, with periodic rotation. Then, 200 ml of each sample were concentrated by precipitation with polyethylene glycol 8000 (10% [wt/vol]; Sigma-Aldrich, Inc.) and NaCl (0.385 M; Sigma-Aldrich, Inc.), as previously J o u r n a l P r e -p r o o f Journal Pre-proof described (Wu et al., 2020) with minor modifications.

Samples with PEG/NaCl were incubated on iced water for 30 min with rotation and centrifuged at 12,000 x g for 1 h, at 4°C. The pellet was resuspended in 1.0 ml of TRIzol reagent (ThermoFisher Scientific) for nucleic acids extraction; 0.2 ml of chloroform was added and mixed by shaking. After a 3 min incubation, samples were centrifuged for 15 min at 12,000 × g at 4°C and the aqueous phase was retained. Then, 4 ul de polyA (High Pure Viral Nucleic Acid Kit, Roche Diagnostics) and 0.5 ml of isopropanol were added to the aqueous phase, incubated for 30 min at -20°C and centrifuged for 10 min at 12000 x g at 4°C.

The pellet was suspended in 0.2 ml of Binding Buffer (High Pure Viral Nucleic Acid Kit, Roche Diagnostics) and nucleic acid extraction was performed with the kit according to the manufacturer's instructions. The nucleic acids were collected in a final volume of 50 ul of elution buffer and stored at -80°C.

The RT-qPCR directed to the N1 genetic region of SARS-CoV-2 was implemented, using the Step Master Mix (ThermoFisher Scientific). Each reaction contained 5 μl of the 50 μl eluted nucleic acids and the reaction volume was adjusted to a final volume of 20 μl with MilliQ water. Thermal cycling (15 min at 50 °C, 20 s at 95 °C followed by 45 cycles of 15 s at 95 (Eurofins) quantified by fluorometric measure (Qubit® 3.0 Fluorometer). NTCs were included in each run. The absolute limit of detection (the lowest viral load that gives 95% of positive results (ALOD95%) was estimated by the Probability of Detection (POD) model (Wilrich and Wilrich, 2009) implemented in the PODLOD calculation program v.9. The absolute limit of quantification (ALOQ) was estimated as the lowest quantity measured with a coefficient of variation (CV) lower than 0.35.

As a human viral indicator, we quantified human polyomavirus (HPyV) in sewage samples to rule out failures in the viral concentration, detection, or other sampling issues such as sewage dilution. Quantification was performed in a subset of the samples including those with nondetectable results for SARS-CoV-2 (n= 62, Table S1 in Supplementary Material). A qPCR assay, previously developed by our workgroup, that simultaneously quantifies JCPyV and Thermal cycling (10 min at 95°C followed by 40 cycles of 15 sec at 95°C, 60 sec at 60°C) was carried out on a Step One Plus (Applied Biosystems) thermocycler. Despite the procedure we applied using TRIzol reagent is indicated for RNA isolation, we were able to co-isolate DNA in the same nucleic acid extracts.

To estimate HPyV concentrations, a standard curve per run was used. It was based on 10-fold serial dilutions (from 1 to 10 5 gc per reaction) of a plasmid DNA containing a JCPyV VP2-VP1 sequence (subtype 2 A), quantified by fluorometric measure (Qubit® 3.0 Fluorometer).

The concentration of the bovine coronavirus (BCoV), Mebus strain, was determined by RT-qPCR as described previously (Cho et al., 2010) in the concentrated samples as well as in the initial sample. Briefly, real-time TaqMan PCR assays were performed using the TaqMan™ 

To characterize the concentration process, we estimated the recovery of SARS-CoV-2 and a bovine coronavirus, Mebus strain, that could behave as a potential surrogate. Viral recovery (VR) of both viruses was calculated as the ratio of the total equivalent genome copies (eq gc) in the viral concentrate and the total eq gc in the wastewater sample:

In a preliminary assay, BCoV Mebus recovery and its associated variability were calculated in quadruplicate from one sample. Briefly, 800 ml of a wastewater sample belonging to the SP location were spiked with the BCoV Mebus strain before virus concentration, achieving an initial concentration of 2,5×10 2 UFF/ml. Then, the sample was homogenized and processed in four parallel subsamples of 200 ml. Also, to assess VR variability across different samples, BCoV was used to spike additional wastewater samples (n=15) and viral J o u r n a l P r e -p r o o f recovery was also calculated. VR of SARS-CoV-2 was evaluated using non-spiked wastewater samples from point SP. Selected samples (n = 7) were those in which the virus was in a concentration high enough, quantifiable in the wastewater sample before the concentration process. Also, to describe viral recovery distributions of SARS-CoV-2 and BCoV, data that could be used in future prevalence studies, the shape parameters of a beta distribution were estimated using the ebeta function from the EnvStats package. Monte Carlo simulations were performed to obtain a random sample (n=10,000) of the estimated distributions of both viruses. Beta distributions were then compared using the Two-sample Kolmogorov-Smirnov test (stats package).

Sample limit of detection and quantification was calculated as described previously (Rajal et al., 2007) where C sample is the genomic copies of the qPCR reaction, V na,qPCR is the volume of nucleic acids in the qPCR reaction, V f,ext is the final volume of the nucleic acids extraction, V i,ext is the concentrated sample volume used for nucleic acids extraction, V conc is the volume of the viral concentrate, V sample is the volume of the wastewater sample and VR is the mean of the viral recovery based on the SARS-CoV-2 estimation. All volumes were expressed in ml.

A non-parametric Wilcoxon rank-sum test (stats) was used to compare SARS-CoV-2 and HPyV concentrations in wastewater and correlation analyses were performed using the Spearman test package using RStudio v1.4.1106.

COVID-19 cases. However, this correlation analysis could not be carried out for the neighborhoods where SP and SF are located, given the lack of epidemiological data.

In the correlation analyses samples where SARS-CoV-2 was detected below the ALOQ, a value of half of the SLOQ (median) was used. Instead, for those samples where SARS-CoV-2 could not be detected or was detected below the ALOD95%, half of the SLOD (median) was used.

Viruses present in wastewater samples were concentrated by precipitation with polyethylene for enveloped viruses such as murine coronavirus. Therefore, we concentrated the entire matrix of the wastewater samples without prior clarification and extracted the viral nucleic acids from the pellet with several purification steps to remove the qPCR inhibitors.

To characterize the whole procedure from viral concentration to quantification, we determined viral recovery for SARS-CoV-2 and a potential surrogate virus: a bovine coronavirus, Mebus strain (Table 1) . Since viral recovery is highly variable, we compared its performance within and between samples using BCoV Mebus. In a preliminary assay, viral

recovery in a single sample tested in quadruplicate was evaluated. Because of its high variability, viral recovery is often better represented by beta distributions, data that can also be used in future prevalence studies. Empiric data was then used to estimate shape parameters of a beta distribution to represent viral recovery. Distributions were then simulated (10000 mc simulations) and their corresponding statistics were obtained (Table 1) .

Although empiric recovery data were not significantly different between SARS-CoV-2 and BCoV Mebus, the corresponding estimated beta distributions did differ (p-value = 2.2 x 10 -16 ).

SARS-CoV-2 genome was detected and quantified by RT-qPCR using CDC N1 primers and probe set (Lu et al., 2020) which was originally designed for clinical samples. In the wastewater samples studied in this work, the ALOD95% of the implemented technique was estimated as 2.6 gc per reaction, (95% CI = 1.3-5.2). The limit of quantification (ALOQ) has been estimated as 10.0 gc per reaction (CV = 28.4%).

The sample limit of detection (SLOD) is the minimum viral concentration in the original J o u r n a l P r e -p r o o f sample that could be detected after the concentration procedure. It depends not only on the sensitivity of the RT-qPCR but also on the variables of the whole molecular detection and the characteristics of the samples that affect the concentration process. Considering an RT-qPCR ALOD of 2.6 gc and a median viral recovery for SARS-CoV-2 of 0.085, the sample limit of detection of this procedure was 1.53 gc/ml with a 95% CI of 0.73-6.19 gc/ml which reflects dispersion in the viral recovery factor. although not statistically significant. This could be explained either by the concentration levels or by the ratio of RNA / viral particles, which could be higher for BCoV.

As another checkpoint, we quantified human polyomaviruses JCPyV and BKPyV, as viral indicators, associated with human fecal/urine contamination, especially in those samples where SARS-CoV-2 was not detected. Detection and quantification of these viruses' load (JC and BK in this work) can account for sewage dilutions due to runoffs, rain and pipe infiltrations that can occur along the sewage system. These variations can also cause fluctuation in SARS-CoV-2 loads in sewage besides COVID-19 cases in the community.

Also, they can account for failures in the viral concentration or detection inhibitions. In this study, these viral controls performed well for the warning of such events.

As the main goal of this survey, the concentration of SARS-CoV-2 genome in wastewater was measured and it ranged from 10 -1 to 10 3 gc/ml, values within the range reported by other J o u r n a l P r e -p r o o f authors (Ahmed et al., 2020a; Randazzo et al., 2020; Wu et al., 2020; Wurtzer et al., 2020) .

This broad range is comprehensible since it depends not only on the epidemiological situation of the outbreak but also on the sampled sewer system, type of collection site, viral recovery of the concentration method. Then, in addition to a real difference in viral excretion among the populations, dilution effects on the wastewaters due to the commercial activities or run-off water, or a non-characterized negative effect on the viral concentration/detection processes cannot be ruled out.

In this study, more consistent results were obtained when samples were collected from the inlet pipes of the WWTP than to the manhole. The waste homogenization obtained in these downstream collection points could have compensated for the lack of a composite sample, which may not occur when moving upstream to collect samples from the manholes.

Unfortunately, epidemiological information representing the small communities of SP and SF neighborhoods was not available for this study. Region. Given that the two sampling points in this district were closely located, the movement of individuals between them for commercial and industrial purposes and medical assistance can justify that the region behaves epidemiologically better as a unit, which is supported by the correlations that were stronger for the entire district than for the individual localities (ρ = 0,795 to 0,812 vs ρ = 0,443 to 0,727).

In this study, the best correlation between the viral RNA concentrations in wastewater and the clinical data was obtained for a 20-day span time of reported COVID-19 cases. A long duration of viral shedding in feces has been informed, even for several weeks, however, the integrity and thus, the detection of viral genomes in wastewater matrices can be altered over time. The decay of the viral RNA excreted in wastewater was shown to vary in a wide range (Ahmed et al., 2020b) . Then, the appropriate period of accumulated cases to evaluate in these studies should be a balance between the duration of fecal viral excretion and the decay of viral RNA in these environments. According to our results, the cumulative number of cases in periods of 20 days correlates strongly with the viral RNA quantified in wastewaters (ρ = 0.812), as other authors also found (Medema et al., 2020b) . This was also observed for shorter periods of 10 and 15 days (ρ = 0.795-0.807). Correlation between daily or weekly COVID-19 cases and SARS-CoV-2 concentration in wastewater per capita per day has also been described recently (Weidhaas et al., 2021) . 

Sewage surveillance of SARS-CoV-2 reflected the population excretion dynamics into the wastewater system. It will be very useful when the number of COVID-19 cases decreases and mass vaccination can be achieved, as it will early warn about an increase in viral circulation in the population. However, differences in the dynamics in different sewer systems can be the results not only of the spread of the virus in the community but also of the intrinsic variables of each sewer. These results can assist local authorities in taking actions of public health, enabling the cooperation between research and management institutions. 

Gastrointestinal Manifestations of SARS-CoV-2 Infection and Virus Load in Fecal Samples From a Hong Kong Cohort: Systematic Review and Meta-analysis

Development of a panel of multiplex real-time polymerase chain reaction assays for simultaneous detection of major agents causing calf diarrhea in feces

Ensure availability and sustainable management of water and sanitation for all in Latin America and the Caribbean

SARS-CoV-2 from faeces to wastewater treatment: What do we know? A review

Evaluating recovery, cost, and throughput of different concentration methods for SARS-CoV-2 wastewater-based epidemiology

US CDC real-time reverse transcription PCR panel for detection of severe acute respiratory syndrome Coronavirus 2

Implementation of environmental surveillance for SARS-CoV-2 virus to support public health decisions: Opportunities and challenges

Presence of SARS-Coronavirus-2 RNA in Sewage and Correlation with

Prevalence in the Early Stage of the Epidemic in the Netherlands

SARS-CoV-2 : sewage surveillance as an early warning system and challenges in developing countries

Wastewater-based epidemiology as a useful tool to track SARS-CoV-2 and support public health policies at municipal level in Brazil

Validation of hollow fiber ultrafiltration and real-time PCR using bacteriophage PP7 as surrogate for the quantification of viruses from water samples

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

Systematic review with meta-analysis: SARS-CoV-2 stool testing and the potential for faecal-oral transmission

Correlation of SARS-CoV-2 RNA

Estimation of the POD function and the LOD of a qualitative microbiological measurement method

Virological assessment of hospitalized patients with COVID-2019

SARS-CoV-2 Titers in Wastewater Are Higher than Expected

Evaluation of lockdown effect on SARS-CoV-2 dynamics through viral genome quantification in waste water

Surveillance of adenoviruses and noroviruses in European recreational waters

Survivability, Partitioning, and Recovery of Enveloped Viruses in Untreated Municipal Wastewater