Sewage surveillance for SARS-CoV-2: Molecular detection, quantification, and normalization factors.

The presence of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in wastewater systems provides a primary indication of the coronavirus disease 2019 (COVID-19) spread throughout communities worldwide. Droplet digital polymerase chain reaction (dd-PCR) or reverse transcription-polymerase chain reaction (RT-PCR) administration of SARS-CoV-2 in wastewaters provides a reliable and efficient technology for gathering secondary local-level public health data. Often the accuracy of prevalence estimation is hampered by many methodological issues connected with wastewater surveillance. Still, more studies are needed to use and create efficient approaches for deciphering the actual SARS-CoV-2 indication from noise in the specimens/samples. Nearly 39-65% of positive patients and asymptomatic carriers expel the virus through their faeces however, only ∼6% of the infected hosts eject it through their urine. COVID-19 positive patients can shed the remnants of the SARS-CoV-2 RNA virus within the concentrations ∼103-108 copies/L. However, it can decrease up to 102 copies/L in wastewaters due to dilution. Environmental virology and microbiology laboratories play a significant role in the identification and analysis of SARS-CoV-2 ribonucleic acid (RNA) in waste and ambient waters worldwide. Virus extraction or recovery from the wastewater (However, due to lack of knowledge, established procedures, and integrated quality assurance/quality control (QA/QC) approaches, the novel coronavirus RNA investigation for estimating current illnesses and predicting future outbreaks is insufficient and/or conducted inadequately. The present manuscript is a technical review of the various methods and factors considered during the identification of SARS-CoV-2 genetic material in wastewaters and/or sludge, including tips and tricks to be taken care of during sampling, virus concentration, normalization, PCR inhibition, and trend line smoothening when compared with clinically active/positive cases.

Introduction

As of November 2021, more than 262 million confirmed cases of COVID-19 have appeared worldwide, triggering the ongoing pandemic. Wastewater-based epidemiology (WBE) has proved to be a swift and supplementary tool for health authorities worldwide to monitor COVID-19 and also track the newly emerging variants of concern. The use of WBE has previously been effective in monitoring a range of harmful viruses including polio, Noro, dengue [1], and recently SARS-CoV throughout the world [2, 3∗, 4]. It targets the residues of the viruses' DNA/RNA that function as the pathogens' population quantitation. Unlike other respiratory viruses, the SARS-CoV-2 virus is located in most infested people's gastrointestinal tracts and faeces [5]. Its injection into the wastewater streams was found in the initial pandemic phases, even prior to the first case in the communities were discovered [6]. WBE, therefore, holds the capability of acting as an early warning system for the development of COVID-19 at a societal stage by analyzing the signals of the viral load in wastewater samples pooled by population [7].

It is worth noting that changes in wastewater SARS-CoV-2 RNA content are not necessarily proportionate to changes in confirmed cases or incidences [8, 9, 10, 11]. This data implies that wastewater is a complicated matrix because of the variability in numerous aspects including volume and duration of individual virus shedding, rates of RNA degradation, and carrier movement. Also, clinically confirmed cases do not cover the entire population of infections but only in the tested population. Confirmed cases are dependent on a scale of clinical testing (more testing finds more infections) and selection of examinee groups (generally, the positive ratio decreases when examinee is selected randomly or an examinee group becomes larger). These fluctuations in confirmed cases also cause a gap between confirmed cases and wastewater epidemic data, as well as catchment features and experimental errors. Therefore, more profound knowledge is needed about the SARS-CoV-2 virus variability in wastewaters and how that corresponds to the actual occurrence or dominance of COVID-19 in the conducive population to ensure relevance of the COVID-19 WBE for public health policymaking. Inconsistencies in sample collection and processing in laboratories, degradation of nucleic acid owing to transit time and sewer environments, and dilution of signal due to fluctuations in precipitation and daytime flow are all measures of target signal inconsistency in wastewaters [12, 13, 14].

To assess the congruency of viral RNA content in wastewater with clinical cases, [15] built a model that included additional factors of high ambiguity and variability, such as rate of flow and per capita wastewater output, and viral rate of shedding. Several normalizing strategies are studied to adjust for the variation in faecal material induced by dilution, primarily using the two most familiar human faecal viral indicators (i.e., cross-assembly phage (crAssphage) and pepper mild mottle virus (PMMoV)) [16]. There are several limiting conditions in the detection of the novel coronavirus in wastewater systems. Municipal wastewater is a complicated and unpredictable concoction encompassing innumerable microorganisms and latent inhibitors along with several variants/strains of the SARS-CoV-2 virus enhancing the pre-requisites for precise procedures. In addition, SARS-CoV-2 genetic material appears in wastewater in minuscule requiring the design of trials with lower detection limits [17]. Thus, standardization of protocols and stringent methodologies will decrease the probability of errors (false positive, false negative) and thereby aid public health decision-making in reducing COVID-19 epidemics.

Here we review the recent literature on concentration and identification methods of SARS-CoV-2 viral RNA from wastewater to assess local COVID-19 outbreaks including PCR technology, PCR inhibition, and virus normalization factors for WBE of SARS-CoV-2. In addition, we discuss the future implications of SARS-CoV-2 monitoring in wastewater; and how the use of vaccines may help to end the pandemic.

Virus extraction from wastewater sample(s)

To get detectable levels of viral nucleic acid, a large volume sampling of more than 1 L is required for the concentration step [18]. The sample technique and time of sampling are critical elements for using WBE, since they can affect data interpretation and possible cross-study comparisons [19]. Majority of research have concentrated on water samples, both small and large grab samples [20, 21, 22] as well as time or flow proportional composite samples [22, 23, 24]. Other research, on the other hand, have used techniques such as Moore swabs, a gauze pad suspended in flowing WW and then processed. Grabs may be less expensive and easier to conduct than composite samples, but they may also have a higher level of unpredictability. This variability is primarily determined by the volumes utilized, the distance of WWTP and sewer due to the viable virus decay over time, the time of day selected, given variations in both water consumption and source strength, which are connected to bathroom habits. When infection frequency is little and/or the sampled population size is small, such as in near-source sampling, when samples are gathered upstream in the sewage network close to the discharge source, quantitative estimates of the number of people affected are likely to be difficult to come by (e.g., outside a building). Because contributing events (e.g., toilet flushes) are more discrete and non-aggregated, grab sampling risks missing the event, and composite samples may be severely diluted by analytes lacking wastewater in the latter instance, the likelihood of collecting a representative sample is low.

Several studies have compared the concentration methods for enveloped viruses, notably the novel coronavirus monitoring, in wastewater [25∗∗, 26, 27, 28, 29]. The majority of such studies counted on exogenic viral controls put in wastewater samples to test the effectiveness of the process, and they were complemented by several cautions and restrictions [30]. Technologies that concentrate and quantify SARS-CoV-2 in wastewater have been well studied. These approaches' dependability, repeatability, and sensitivity must be confirmed for a more significant usage of the wastewater data. This “pooling approach” is essential in areas where clinical testing rates for COVID-19 are low, resources are few, or the number of cases is unknown [31]. MS2, a non-enveloped bacteriophage frequently employed as a process control of enteric virus detection, is one of several ways to detect and estimate SARS-CoV-2 concentrations in wastewater (used for process control) [32]. Two more enveloped coronaviruses (BCoV and OC43) are employed further to understand the efficacy in detecting SARS-CoV-2 in wastewater of which, the concentrating pipette (CP)-based concentration approach is more successful than the ViroCap-based (VC-based) when it comes to recovery efficiency and speed. The CP approach is much less time-consuming than the VC-based technique, making the efficacy of MS2 recovery two times greater with the CP technique (53.6%) than with the VC-based method (24.7%) [8]. When it came to retrieving encapsulated coronaviruses from wastewater, V.C. was less effective (BCoV: 7.2%). Solids are eliminated prior to viral concentration, which might be an issue with this quick CP approach, since enveloped viruses have been discovered as increasingly connected to the surface of the particles present in wastewater compared to viruses that lack envelopes, in prior research [33]. Virus recovery efficiency post fractionation, polyethylene glycol (PEG) precipitation, and viral RNA extraction in vesicular stomatitis virus (VSV)-spiked wastewater samples were done by Ref. [34] (Equation (1)).

[35] devised the 4S method to extract SARS-CoV-2 RNA from wastewater using a vacuum column to concentrate and purify RNA in less than 3 h. This approach helps researchers at the University of California, Berkeley, to detect SARS-CoV-2 envelope (E) and nucleocapsid (N) gene RNA and PMMoV RNA using RT-qPCR probe-mediated detection. Other existing concentration methods use chemical precipitation [36], size exclusion [37], adsorption through membranes [38], ultracentrifugation [39], flocculation [26], or a combination of such technologies [40,41] with wastewater input amounts ranging from 15 to 250 mL. The approach described by Ref. [42], in which the amount of RNA of SARS-CoV-2 conducted from wastewater of 45 L quantity via electropositive cartridges, is the only large-volume concentration technique published in previous articles; nonetheless, being very labour-intensive. Table 1 shows the various approaches by researchers reported recently on the wastewater monitoring methods of SARS-CoV-2.

Table 1

Recent studies on SARS-CoV-2 sampling, sample pre-treatment, RNA gene detection, and sequencing for COVID-19 surveillance via non-clinical approach.

sample	Region	Population size	Sampling type	Time period	Virus filtration and concentration	Recovery efficiency/Normalization biomarkers	Detection method and target gene(s)	Internal control/PCR inhibition	Sequencing/variant analysis (+/−)	Reference
Pumping station (1) and Wastewater treatment plants (2)	Southeast Queensland, Australia	42,612–1,106,892	Grab sampling and automated sampling (conventional refrigerated autosampler and in-situ high-frequency autosampler)	January 2020–April 2020	Direct RNA extraction from electronegative membranes and ultrafiltration	–	RT-qPCR,	Oncorhynchus keta (O. keta)- copy number 104/reaction	+ (Sanger and MiSeq, Illumina)	Ahmed et al., 2020a [15]
Wastewater treatment plants (9)	Central Ohio, US	14,000–49,000	24-h composite samples-twice a week	July 2020–January 2021	Adsorption-precipitation by a positively charged filter followed by flocculation and centrifugal ultrafiltration	Spiking surrogates: male-specific coliphage MS2 (ATCC cat. No. 15597-B1), bovine coronavirus (BCoV strain ML-6 mebus), and human coronavirus OC43 (ATCC cat. No. VR-1558)	dd-PCR, N-gene and E-gene	Firefly (Coleoptera) Luciferase control RNA	+ (Next generation sequencing)	Ai et al., 2021 [8]
Wastewater treatment facilities (6)	California–San Francisco Bay area, US	82,818–1,500,000	24-h time-weighted composite samples-weekly	April 2020–September 2020	Modified 4S method	crAssphage CPQ_056, pepper mild mottle virus coat protein gene (PMMoV), Bacteroides 16S ribosomal RNA HF183/BacR287, bovine coronavirus transmembrane protein gene (BCoV), Synthetic Oligomer Construct T33-21 free-RNA (SOC), and human 18S rRNA	RT-qPCR, N1 gene	VetMAX™ Xeno™ Internal positive control (Xeno)	–	Greenwald et al., 2021 [61∗]
Wastewater treatment plants (5)	Ishikawa and Toyama Prefecture, Japan	31,501 - 233,480	Grab sampling	March 2020–May 2020	PEG precipitation and quantification of F phage	–	qRT-PCR, CDCN2, CDCN3, and NIID assays	Murine norovirus	+ (Sanger sequencing)	Hata et al., 2021
Wastewater treatment plants	Amsterdam and Utrecht, Netherlands	267,900 - 669,400	24 h composite sampling	March 2020–March 2021	Centrifugation and ultrafiltration	CrAssphage CPQ_064	RT-qPCR and RT-ddPCR, N2 assay	Mouse Hepatitis Virus (MHV)-A59	+ (Cell culture and whole genome sequencing)	Heijnen et al., 2021
Wastewater treatment plant (1)	Ahmedabad, Gujarat, India	7,800,000	Grab sampling	May 2020	PEG precipitation	–	RT-qPCR, ORF1ab, N and S gene	Bacteriophage MS2	–	Kumar et al., 2020 [17]
Wastewater treatment plants (3)	Milan, Italy	900,000–1,050,000	24 h composite sampling	February 2020–April 2020	PEG-dextran method	–	Nested RT-PCR and RT-qPCR, ORF1ab gene	OneStep PCR Inhibitor Removal Kit	+	La Rosa et al., 2020 [3∗]
Lift stations (7) and Wastewater treatment plants (15)	Vancouver, Edmonton, Toronto, Montreal, Halifax, and Northwest Territories of Canada	87,89,211	24 h composite sampling	February 2021–March 2021	Magna Pure 96 DNA and Viral NA Large Volume Kit	–	RT-qPCR, N-gene, and S gene	–	+ (Sdel and SN501Y assays)	Peterson et al., 2021
Wastewater treatment plants (3)- raw and treated wastewater	Milano and Monza e Brianza, Italy	200,0000	Grab sampling	April 2020	Whatman GF/F and nitrocellulose Millipore MCE filters	Caffeine quantification	RT-qPCR, N, ORF1ab and E gene	QIAMP Viral RNA mini kit	+ (Ion Torrent PGM sequencer)	Rimoldi et al., 2020
Manholes and Wastewater treatment plant	Israel	92,406–644,000	24 h composite sewage sample	June 23, 2021	MCE electro-negative membrane	–	RT-qPCR, N-gene (direct RNA extraction- Zymo Research R2042 protocol)	MS2 phage	+ (RT-qPCR primer probes)	Yaniv et al., 2021 [54]

PCR types/techniques for detecting and quantifying SARS-CoV-2

In wastewater matrix/systems, detection of SARS-CoV-2 RNA starts with sampling from inside a wastewater network structure such as wastewater pumping stations, manholes, or an influent from sewage treatment plants (STPs) or close by building outlet discharge. Then, the virus is concentrated, extricated, and the RNA is run through dd-PCR (does not require a standard plot) and/or RT-qPCR (centred on a standard plot) for identification and quantification of SARS-CoV-2 genetic material. Poor viral retrieval and/or testing of minor effective sample volume (ESV), inefficiency of extracting RNA, intensification-inhibition in the PCR assay, and inadequate sensitivity assessment are all issues that affect sample processing and analysis [41,43]. A number of strategies for removing or inactivating PCR inhibiting entities have been documented, although none of them are always efficient [44]. Process controls, also known as external/internal controls, should be used to evaluate the efficiency of methods. According to the materials used and the points at which the controls are spiked into the sample, there are mainly three types of controls: (1) whole process controls [45,46], which are added before viruses are concentrated from the sample; (2) molecular process controls [47, 48∗, 49], which are added before nucleic acid extractions; and (3) RT-qPCR controls [50, 51, 52∗], which are added before the PCR processes. The impacts of inhibitors are reflected in the recovery yields of the process controls, which helps to better comprehend the study's conclusions. Many prior studies have attempted to establish a threshold value for the projected recovery of process controls in order to confirm the accuracy of viral detection; however, as shown below, there is presently no consensus on that number [52].

dd-PCR is considerably vigorous in managing PCR inhibition conditions than standard quantitative RT-qPCR assays. The Qiagen kit operated for extracting RNA contains many inhibitors elimination processes [53]. SARS-CoV-2 variants can be detected via PCR and/or Next-generation sequencing (NGS) technique [54,55]. designed direct qPCR probes and primers on the basis of variations in sequences between the newly emerging strains (Alpha (B.1.1.7), Beta (B.1.351), Gamma variant (P.1), and Delta variant (B.1.617) with the wild type SARS-CoV-2. The novel primers for detecting P.1 and B.1.617 variants had shown to be both specific and sensitive, with a limit of detection (LOD) 10° in both sterile conditions and sewage environment systems. Although dd-PCR has not been extensively used, it has been evaluated and proved helpful, mainly when viral loads are low, as they are during the decline in the virus load and/or in the course of early phases of virus dissemination [56]. As a result, more research is needed to determine the efficacy of RT-ddPCR.

Normalization

Normalization is recently considered effective since viral concentration in wastewater is affected by dilution with non-faecal wastewater and quantity of faecal discharge at the corresponding time of sampling. Normalization concepts using biomarkers, higher frequency of sampling, sample replicate processing, and smoothening/forecasting have been used to address some of the drivers of variability in SARS-CoV-2 identification. Normalization with population, the target signal, or an endogenic biomarker, enables to minimize inconsistency in data and scale estimates for cross-sample and cross-location comparisons. To compute the per capita load [57] or a chemical parameter, standardized concentrations of wastewaters to population and flow rate across WBE experiments (e.g., caffeine) are used [34]. COVID-19 WBE assays typically rely on the detection of a non-enveloped plant virus viz., PMMoV RNA detection. However, its quantities in sewage fluctuate depending on the season and local food [34]. A non-enveloped DNA virus viz., cross-assembly phage (crAssphage), that contaminates the Bacteroides (commensal bacterium) present in human gut, is another normalization biomarker that has been widely put to use [58]. However, crAssphage shedding per person varies highly, this normalization technique can be applied for population size >5000. Furthermore, the 16S rRNA gene of the human-specific HF183 Bacteroides is commonly utilized to identify faecal pollution in natural waters, and subsequent research [34] have focused on HF183 rRNA to improve the sensitivity of the assay. Finally, since it aims at human cells excreted through faeces, the 18S rRNA, ribosomal subunit (18S) present in humans was recommended as the biomarker for normalization [35]. Figure 1 shows the various steps for sample collection, virus concentration, PCR standards, normalization of data, and sequencing for variant identification, all of which need to be taken extreme care of for WBE of SARS-CoV-2. Moreover, flow normalization method (with electrical conductivity (EC) check) is economic and reliable [59] as shown in equations (2) and (3).

Figure 1

Wastewater sampling, virus concentration, internal controls, normalization of SARS-CoV-2 signal and variant detection for efficient public health-related decision-making, and evaluation of vaccination efficiency against SARS-CoV-2 variants.

Also, the number of people in a catchment can be back-calculated with the help of tracing biomarkers [60].where, P = population/number of people in the watershed, C  = concentration of the biomarker used (g/L), Q = flow volume (L/d), and f  = specific biomarker load (g/P/d).

Also, three equations to find the percentage of domestic wastewater using flow data (equation (5)), EC data (equation (6)), and crAssphage data (equation (7)) are given as follows: where, DDWF = average daily domestic wastewater generation by each person, V24h = wastewater volume measured over 24 h sampling time, V  = average wastewater volume during dry weather flow (40th percentile of daily 261 volumes derived from 1-year time series of flow monitoring), EC  = measured EC values in samples, EC  = average EC of all samples whose flow ranges between the 10th and 50th percentile of daily volumes derived from 1-year time series of flow monitoring, CrAss  = genome copies of the phage/ml sample, and CrAss  = average CrAss of all samples with flows ranging between 10th and 50th percentile of daily volumes derived from a 1-year time series of flow monitoring (genome copies/ml sample).

[8] conducted wastewater surveillance of SARS-CoV-2, where the measurements were taken from the influent wastewater samples of the capital and seven other cities of varying sizes in central Ohio for three target genes (i.e., E gene regions, N1 gene, and N2 gene) in a period ranging from July 2020 to January 2021. Quantification of crAssphage and PMMoV (human faecal viruses) was carried out, normalizing the detected SARS-CoV-2 gene concentrations in the sample to regulate human-specific faecal strength better. This followed normalization of the RNA intensities using PMMoV RNA (concentration (average) throughout the samples. Visually, both methods had increased the concordance of viral concentrations and case numbers but not statistically (Spearman rank correlation). Therefore, interestingly neither the PMMoV nor the crAssphage normalization procedures significantly improved the association with new case numbers or the estimate models. In yet another research carried out by Ref. [61], four normalization indicators were used to account for variance in wastewater sample faecal content viz., PMMoV, crAssphage, 18S rRNA (human), and Bacteroides ribosomal RNA. CrAssphage had the slightest geographical and temporal changeability of the group. Both normalized to crAssphage and unnormalized data signals of SARS-CoV-2 RNA peaks conclusively and notably correlated (Kendall's Tau-b (τ) = 0.43 and 0.38, respectively) with the clinical tested confirmed COVID-19 case data trends.

Similar to normalizing the target signal, smoothing processes can aid in determining sequential patterns in the incidence of COVID-19. Although 7-day moving averages are frequently utilized to examine real-time clinical data patterns [62], low frequency of wastewater sample collection (e.g., once or twice a week) makes it difficult to use such an approach. As a result, smoothing approaches such as locally weighted scatter plot smoothing, which may be employed to lesser sampling frequency data and minimalize the damage to temporal resolution, are required [63]. Nevertheless, there is no benchmark value for the bandwidth variable (similar to choosing a gap of 7 days for moving averages). Additionally, in research including lowess, the bandwidth selection technique is rarely mentioned [31,64].

The detection boundary is determined by the approaches employed to collect hereditary matter and the scope of local clinical examining, and it might necessitate sewer shed-specific evaluation. With clinical testing capabilities, this value may change over time. A methodical technique to estimate this amount across researches can help understand non-detects and predict the per capita quantum of SARS-CoV-2 positive cases over which WBE (of SARS-CoV-2) becomes an effective public health monitoring tool [61]. Recent research used a dataset of 1687 samples to establish an evident WBE case detection boundary that was huge enough to allow recurrent wastewater assessments at small case numbers [65]. Researchers have calculated this value observationally using fewer data points by reporting the number of instances they could identify or quantify [66].

Future implications for the WBE tool

Over a dozen investigations worldwide demonstrate a shift toward using wastewater inspection to estimate SARS-CoV-2 transmission in different locations. SARS-CoV-2 identification studies mostly show the slope of the graph plotted between the cumulative infections’ cases and the number of copies of genes, post normalization in the population living in the catchment area. This slope can be linked to the number of new cases on average over the sample period. As a result of the association, qPCR data may forecast future contagions. Because wastewater sampling captures all infections, whether symptomatic or not, a lag time enhanced the association between the average number of new cases and the slope. SARS-CoV-2 variants (Alpha (B.1.1.7), Beta (B.1.351), Gamma variant (P.1), and Delta variant (B.1.617)) can also be detected via wastewater surveillance, and the emergence of new variants can be reported while mitigation strategies for future outbreaks in targeted regions can be focused.

[55] created a RT-qPCR technique for the fast, sensitive, and direct identification of SARS-CoV-2 P.1 and B.1.617 variants. WBE can be a valuable instrument for explaining population diversity, immune response, and monitoring and revealing vaccination attempts and dynamics of individual resilience to SARS-CoV-2 infection in the community. Wastewater should play a major part in the surveillance of a variety of different infectious illnesses in the future [67]. developed connections linking SARS-CoV-2 genetic material load in wastewater and pandemic health indices using modelling approaches such as distributed/fixed lag modelling, linear regression, and artificial neural networks. Such models [68,69] can be very helpful in conducting risk assessments for future outbreaks.

Conclusion

Both methodological studies for testing SARS-CoV-2 in wastewaters and the effectuation of WBE happened simultaneously during the COVID-19 epidemic. Studies show that patterns in wastewater monitoring figures match COVID-19 incidence trends; they also offer strategies for making wastewater signals more interpretable and comparable between studies. Clinical and environmental monitoring data may be integrated to construct robust models that can be used to examine the ongoing COVID-19 infection dynamics and give an early warning system for increasing hospital admissions. To be relevant for decision-making for public health, COVID-19 WBE must be sure that the resultant signal of SARS-CoV-2 accurately represents the trends of COVID-19 in the causative population. Data such as rigorous QA/QC courses, characteristic sampling tactics, efficacious virus concentration techniques and effectual RNA elution and recovery methods, accurate assessment of PCR inhibition, the incorporation of controls for processing of wastewater/sludge samples, significances for dd-PCR/RT-PCR assay availability, selection and scientific evaluation of signal/trend, etc., are all recommended to reduce incorrect positive and negative cases in SARS-CoV-2 surveillance.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.