key: cord-0906837-uuzz4ij1
authors: Zheng, Xiawan; Deng, Yu; Xu, Xiaoqing; Li, Shuxian; Zhang, Yulin; Ding, Jiahui; On, Hei Yin; Lai, Jimmy C.C.; In Yau, Chung; Chin, Alex W.H.; Poon, Leo L.M.; Tun, Hein M.; Zhang, Tong
title: Comparison of virus concentration methods and RNA extraction methods for SARS-CoV-2 wastewater surveillance
date: 2022-02-05
journal: Sci Total Environ
DOI: 10.1016/j.scitotenv.2022.153687
sha: 7ab067db739885e9d00c392e73ef77bee05d89d4
doc_id: 906837
cord_uid: uuzz4ij1

Wastewater surveillance is a promising tool for population-level monitoring of the spread of infectious diseases, such as the coronavirus disease 2019 (COVID-19). Different from clinical specimens, viruses in community-scale wastewater samples need to be concentrated before detection because viral RNA is highly diluted. The present study evaluated eleven different virus concentration methods for the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in wastewater. First, eight concentration methods of different principles were compared using spiked wastewater at a starting volume of 30 mL. Ultracentrifugation was the most effective method with a viral recovery efficiency of 25 ± 6%. The second-best option, AlCl3 precipitation method, yielded a lower recovery efficiency, only approximately half that of the ultracentrifugation method. Second, the potential of increasing the sensitivity of the method was explored using three concentration methods starting with a larger volume of 1000 mL. Although ultracentrifugation using a large volume outperformed the other two large-volume methods, it only yielded a comparable method sensitivity as the ultracentrifugation using a small volume (30 mL). Thus, ultracentrifugation using less volume of wastewater is more preferable considering the sampling processing throughput. Third, a comparison of two viral RNA extraction methods showed that the lysis-buffer-based extraction method resulted in higher viral recovery efficiencies, with cycle threshold (Ct) values 0.9–4.2 lower than those obtained for the acid-guanidinium-phenol-based method using spiked samples. These results were further confirmed by using positive wastewater samples concentrated by ultracentrifugation and extracted separately by the two viral RNA extraction methods. In summary, concentration using ultracentrifugation followed by the lysis buffer-based extraction method enables sensitive and robust detection of SARS-CoV-2 for wastewater surveillance.

respiratory syndrome coronavirus 2 (SARS-CoV-2) in wastewater. First, eight concentration methods of different principles were compared using spiked wastewater at a starting volume of 30 mL. Ultracentrifugation was the most effective method with a viral recovery efficiency of 25 ± 6%. The second-best option, AlCl 3 precipitation method, yielded a lower recovery efficiency, only approximately half that of the ultracentrifugation method. Second, the potential of increasing the sensitivity of the method was explored using three concentration methods starting with a larger volume of 1,000 mL. Although ultracentrifugation using a large volume outperformed the other two large-volume methods, it only yielded a comparable method sensitivity as the ultracentrifugation using a small volume (30 mL). Thus, ultracentrifugation using less volume of wastewater is more preferable considering the sampling processing throughput.

Third, a comparison of two viral RNA extraction methods showed that the lysis-bufferbased extraction method resulted in higher viral recovery efficiencies, with cycle threshold (Ct) values 0.9 -4.2 lower than those obtained for the acid-guanidiniumphenol-based method using spiked samples. These results were further confirmed by

The coronavirus disease 2019 outbreak caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection has severely affected the global public health and economy. During the onset of the pandemic, increasing screening and testing is important for the timely implementation of control policies (Mercer and Salit, 2021) . However, screening and testing have challenges, particularly when the clinical testing capacity is underutilized or unavailable. The emergence of new strains, such as alpha, beta, delta, and omicron, shows rapid viral evolution and challenges community-level testing and control of the pandemic (Hrudey and Conant, 2021) .

Wastewater-based epidemiology (WBE) tool has been effective for community monitoring of gastrointestinal pathogenic viruses, such as poliovirus, norovirus, and hepatitis A virus (Brouwer et al., 2018; Hellmer et al., 2014) . Recent studies have indicated that SARS-CoV-2 RNA fragments can be detected in fecal samples of patients of COVID-19, with a viral load ranging 2 -8 log 10 copies/mL (Cheung et al., 2020; Guo et al., 2021; Wolfel et al., 2020; Zheng et al., 2020) . The application of the WBE tool in SARS-CoV-2 detection is effective and efficient, providing early warning signals before community outbreaks (Medema et al., 2020; Nemudryi et al., 2020; Xu et al., 2021) .

Furthermore, it enables the monitoring of the progression of new strains ahead of clinical testing (Graber et al., 2021; Jahn et al., 2021) . In addition, it can serve as an unbiased and non-invasive surveillance tool for identifying previously unknown symptomatic or asymptomatic patients (Mallapaty, 2020; Wu et al., 2020) . Effective and robust wastewater testing for SARS-CoV-2 can help provide important and timely information J o u r n a l P r e -p r o o f Journal Pre-proof to stakeholders and decision-makers for making prompt control measures to fight COVID-19, such as evaluation of the lockdown effect on pandemic dynamics (Hillary et al., 2021; Wurtzer et al., 2020) , uncovering hidden cases on college campuses (Gibas et al., 2021) , and assessment of the effectiveness of vaccination (Bivins and Bibby, 2021) .

Compared with the clinical testing, the detection of SARS-CoV-2 in wastewater is difficult due to low viral amounts and matrix effects on detection imposed by the complex components in wastewater. Therefore, effective virus concentration methods are essential for the sensitive detection of low concentration SARS-CoV-2 in wastewater. Up to now, researchers adopted various different concentration methods, such as polyethylene glycol (PEG) precipitation (Wu et al., 2020) , AlCl 3 precipitation (Randazzo et al., 2020) , ultrafiltration with centrifugal filters (Medema et al., 2020) , ultracentrifugation (Wurtzer et al., 2020) , and membrane absorption (Haramoto et al., 2020) .

To date, there are limited systematic comparison studies on concentration methods for SARS-CoV-2 virus in wastewater. Reported studies have focused on the performance of virus concentration methods using different concentration principles (Ahmed et al., 2020; LaTurner et al., 2020) , inter-laboratory comparisons (Pecson et al., 2021) , or different surrogate viruses (Jafferali et al., 2021; Philo et al., 2021) . However, most of the published virus concentration methods are processed with 30 -250 mL wastewater because of the practicality of sample handling in the laboratory and the availability of some special instruments. A few studies have proposed increasing detection sensitivity by increasing wastewater sample volume (Gerrity et al., 2021; McMinn et al., 2021; Xu et al., 2021) . In addition, wastewater characteristics and matrices should be carefully J o u r n a l P r e -p r o o f Journal Pre-proof considered when selecting a virus concentration method for SARS-CoV-2 (Pecson et al., 2021) .

The present study conducted a comprehensive evaluation of the concentration methods commonly used for SARS-CoV-2 in wastewater. Specifically, the objectives of this study were to 1) compare the performance of eight concentration methods using small-volume (30 mL) wastewater, 2) compare the performance of small-volume (30 mL) and largevolume (1000 mL) wastewater concentration methods, and 3) evaluate the effects of two viral RNA extraction methods on SARS-CoV-2 detection.

Twelve raw wastewater samples were collected at the inlets of the Shatin Sewage Wastewater samples were transported to the laboratory on ice and processed within 24 h.

All wastewater samples were heat-inactivated at 60 °C for 30 min to ensure lab safety before sample processing (Chin et al., 2020) . The effect of heat-inactivation on detection and method comparison was not evaluated in this study. Previous studies showed that heat-inactivation at 56 °C for 30 or 60 min could inactivate the infection of the SARS-CoV-2 virus but insignificantly affect the detection of Ct values using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) (Auerswald et al., 2021; Batejat et al., 2021) .

Comparison of different small-volume concentration methods: Eight concentration methods were compared using a starting volume of 30 mL (defined as "small-volume" in the present study). In detail, SARS-CoV-2 virus was collected from the cell culture and (1%, v/v) or 2.5 M MgCl 2 (1%, v/v) were added to wastewater samples. Samples were then passed through an electronegative membrane with 0.45 μm pore size and 47 mm diameter (Merck Millipore) through a filter funnel and a filter flask (Thermo Scientific).

The membrane was removed and eluted with 4 mL of PBS to recover the virus. Viral particles were repeatedly blown and scraped from the membrane using a P200 pipette.

Eluant was further concentrated at 20,000 × g for 2 min at 4 °C to obtain a pellet (approximately 200 μL) for RNA extraction.

Meanwhile, 30 mL of wastewater supernatant without spiking SARS-CoV-2 virus and 30 mL of double-distilled water (ddH 2 O) were concentrated using the eight methods as negative controls. All these wastewater and ddH 2 O samples were finally tested negative.

All the above comparison experiments were conducted for four times using four wastewater samples taken on different dates. All concentrated samples were extracted using QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) following the manufacturer's instructions, with a final elution volume of 40 µL.

Heatinactivated SARS-CoV-2 virus (approximately 10 8 copies/mL) of 1.5 mL were spiked into 4.5 L wastewater supernatant collected after centrifugation at 4,750 × g for 30 min, and mixed completely by shaking for 15 min. Triplicate aliquots of 334 μL heatinactivated SARS-CoV-2 virus was extracted and detected directly to obtain the theoretical spiking virus concentration for calculating the method recovery efficiency in each batch experiment. Then, for large-volume concentration methods, 1,000 mL aliquots were collected and processed separately using one of the following three methods ( (1) centrifugation at 20,000 × g for 30 min and then ultracentrifugation at 150,000 × g for 60 min, (2) AlCl 3 precipitation, and (3) 0.45 μm electronegative membrane absorption.

In addition, 30 mL of the spiked wastewater sample was concentrated by ultracentrifugation at 150,000 × g for 60 min to represent the small-volume method for comparison (Method 1).

Method 9 was based on a combination of centrifugation and ultracentrifugation. First, 1,000 mL spiked wastewater was centrifuged at 20,000 × g for 30 min via the Sorvall LYNX 4000 Superspeed Centrifuge (Thermo Scientific), and approximately 30 mL of the precipitate was retained for further ultracentrifugation at 150,000 × g for 60 min using the Optima XPN Ultracentrifuge (Beckman Coulter). Next, the concentrated viral pellet was resuspended with approximately 400 µL PBS and used for further RNA extraction.

Method 10 was an AlCl 3 precipitation-based concentration method. Ten microliter AlCl 3 solution (1%, 0.3M, v/v) was added to 1,000 mL spiked wastewater to precipitate the viral particles. After shaking at 150 rpm for 30 min and centrifuging at 20,000 × g for 30 min using Sorvall LYNX 4,000 Superspeed Centrifuge (Thermo Scientific), concentrated viral pellets were obtained and used for RNA extraction.

Method 11 was the membrane absorption concentration method. The spiked wastewater (1,000 mL) was passed through a 0.45 μm pore-size, 47 mm diameter electronegative membrane (MF-Millipore, Darmstadt, Germany. Cat. No: HAWP09000). The membrane was collected and washed with 4 mL of PBS to obtain a filter pellet for RNA extraction.

Visible film-forming viral particles can be easily washed off from the membrane in lumps.

Wastewater samples without spiking viruses and ddH 2 O of the same volume were negative controls. No SARS-CoV-2 signal was detected in these negative wastewater or ddH 2 O. The above comparisons of the three large-volume and one small-volume method were conducted three times using wastewater samples taken on different dates. RNA extraction for all the concentrated samples from large-volume and small-volume wastewater were conducted using TRIzol Plus RNA Purification Kit (Thermo Fisher)

with an elution volume of 40 µL, as our pre-experiment showed that the QIAamp Viral RNA Mini Kit did not apply to large-volume wastewater due to spin column clogging issues.

To evaluate the performance of the two viral RNA extraction methods (lysis-buffer-based vs acid-guanidinium-phenol-based), 80 µL of heat-inactivated SARS-CoV-2 (approximately 10 8 copies/mL) was spiked into 240 mL supernatant collected after raw wastewater was centrifugated at 4,750 × g for 30 min, to obtain a final concentration of approximately 10 5 copies/mL. Next, 30 mL samples were concentrated by ultracentrifugation methods at 150,000 × g for 60 min, followed by RNA extraction using the QIAamp Viral RNA Mini Kit (lysis-buffer-based method, called "Viral Kit" in the following description) or TRIzol Plus RNA Purification Kit (acidguanidinium-phenol-based method, called "TRIzol Kit"). For comparison, triplicate samples using wastewater from the same sampling site were used to test the processing variation. The comparison was repeated using wastewater samples from four sampling sites: HP, SSP1, SH, and YT in Hong Kong.

Reagent blanks using 200 µL AVE buffer or RNase-free water from the viral RNA extraction kits were used in each extraction batch as a quality control measure.

J o u r n a l P r e -p r o o f

After centrifugation at 4,750 × g for 30 min, the wastewater supernatant was spiked with inactivated SARS-CoV-2 virus to obtain a concentration of 1 -100 copies/mL wastewater to test the performance of the three methods (Method 1 with Viral Kit or TRIzol Kit, and Method 9 with TRIzol Kit) under the challenge of virus concentration at marginal levels.

To be specific, 400 µL of virus at three concentrations (around 10 3 -10 5 copies/mL) were spiked into 1.2 L wastewater samples to generate spiked wastewater of different concentrations respectively, naming "high-concentration spiked wastewater (approximately 100 copies/mL, called "High" in the following description), "middleconcentration spiked wastewater (approximately 10 copies/mL, called "Medium"), and "low-concentration spiked wastewater (approximately 1 copy/mL, called "Low"). For each virus concentration, two 30 mL spiked samples were concentrated by ultracentrifugation and extracted with two viral extraction kits, that is, TRIzol Kit and Viral Kit. In addition, a 1,000 mL spiked wastewater sample was extracted using the TRIzol Kit after the large-volume concentration methods. Wastewater samples without spiking SARS-CoV-2 virus were used as negative controls. All experiments were performed in duplicate.

The heat-inactivated SARS-CoV-2 with a stock concentration of 6.09 × 10 8 copies/reaction (approximately 10 10 copies/mL virus) was 10-fold diluted using either PBS or negative wastewater samples to obtain a virus concentration range of 6.09 × 10 2 -6.09 × 10 7 copies/reaction (the range of log 10 virus concentration was 3 -8). For each virus concentration, 100 µL of PBS diluted virus or wastewater diluted virus was J o u r n a l P r e -p r o o f extracted using either the TRIzol Kit or Viral Kit, with a final elution of 50 µL. The entire process was repeated in triplicates.

Wastewater samples without spiking were used to evaluate the detection sensitivity of the different concentration methods. The evaluation focused on the comparison of ultracentrifugation with small-volume or large-volume wastewater, and the comparison of extraction by Viral kit or TRIzol kit, based on detection rates and detected Ct values.

These samples were taken from manholes, sewage pumping stations, and wastewater treatment plants in Hong Kong.

RT-qPCR were performed to quantify the extracted viral RNA. For spiked wastewater, the HKU-N probe and primers Similarly, previous studies have used Ct values rather than virus concentrations for comparison (Jafferali et al., 2021; Perez-Cataluna et al., 2021; Philo et al., 2021) .

SARS-CoV-2 virus recovery efficiency was calculated based on the ratio of the detected viral concentration throughout the process and the theoretical spiked virus concentration, which was determined by RT-qPCR after direct RNA extraction. 

One-way analysis of variance (ANOVA) was used to determine the significance of differences between the means of Ct values and recovery efficiency among different concentration methods. If the difference was significant (p < 0.05), a post-hoc t-test was used to determine which methods were significantly different. All analyses were performed using R Studio version 1.3.

The HKU-N and N1 detection assays were used in this study, and they showed similar dynamic linear ranges, with values of 1.056 × 10 -1.056 × 10 7 copies/reaction and 1.052 × 10 -1.052 × 10 7 copies/reaction, respectively. Standard curves of HKU-N and N1 assays showed strong linear fits, with R 2 values ranging from 0.9956 to 0.9996 and 0.9942 to 0.9992, respectively. The Y-intercept of the standard curves for HKU-N and N1 assays ranged from 38.59 to 43.63 and 35.94 to 37.47, respectively. The slopes of the standard curves ranged from -3.295 to -3.487 and -3.111 to -3.427, and the amplification efficiencies ranged from 93.86 to 101.2% and 95.81 to 109.6% for the HKU-N and N1 assays, respectively. Detail information was summarized in Table 1 .

We evaluated the performance of the eight concentrations methods using a small wastewater volume of 30 mL ( Figure 1 ). As shown in Figure 2 , ultracentrifugation had the highest recovery efficiency of 25.4 ± 5.9%, significantly higher than other methods (p J o u r n a l P r e -p r o o f < 0.01), followed by precipitation-based methods, ultrafiltration methods, and finally membrane-absorption methods. The recovery efficiency was similar to those from other studies on the bovine respiratory syncytial virus using ultracentrifugation (Prado et al., 2021) , and higher than the values reported in studies on RNA bacteriophage (Prado et al., 2021) and inactivated SARS-CoV-2 virus (Green et al., 2020) (Table 2 ). The results imply that the method sensitivity of ultracentrifugation outperformed other methods for the detection of SARS-CoV-2. However, ultracentrifugation relies on the ultracentrifugation equipment, not readily available in many laboratories.

The mean recovery efficiency of different precipitants (AlCl 3 , PEG, or MgCl 2 ) showed no significant differences, but AlCl 3 precipitation had the lowest variability. The recovery efficiency of AlCl 3 precipitation (11.0 ± 4.36%) concurred with other studies using the Porcine Epidemic Diarrhea Virus and Mengovirus as the surrogate viruses (Randazzo et al., 2020) (Table 2) . Therefore, AlCl 3 precipitation was considered a preferable substitute method considering the less availability of the special equipment needed in the ultracentrifugation method, although its recovery efficiency was only half of that of the ultracentrifugation method.

The recovery efficiencies of ultrafiltration-based and membrane-absorption methods, ranging from 4 to 11%, were lower than those in other studies (Table 2) . A previous report obtained the highest recovery efficiency using MgCl 2 membrane absorption compared to the other methods for concentrating murine hepatitis virus (MHV) from wastewater influent (Ahmed et al., 2020) , but MgCl 2 membrane absorption showed the most inferior performance for the SARS-CoV-2 virus in this study. Such inconsistency could probably result from the difference in the detachment of viral particles from the J o u r n a l P r e -p r o o f Journal Pre-proof electronegative membrane. Bead beating of the entire membrane was used in the study of Ahmed et al (Ahmed et al., 2020) , while in the present study viral particles were only detached from the membrane by repeated washing and scraping using a P200 pipette.

Except for these operational differences, the differences in matrix components in wastewater samples (Cashdollar and Wymer, 2013) and utilization of different surrogate viruses (MHV vs inactivated SARS-CoV-2) may also contribute to difference. Compared to surrogates, directly spiking with the SARS-CoV-2 virus is more representative to evaluate the performance of virus concentration methods for SARS-CoV-2 wastewater surveillance.

To explore the potential of increasing the method sensitivity by using larger sample volume, three large-volume methods were compared and evaluated. As shown in Figure   3a , large-volume samples concentrated using ultracentrifugation and membrane absorption showed higher recovery efficiencies than using AlCl 3 precipitation. Because the membrane absorption method includes a time-consuming step to filter 1,000 mL of wastewater samples through the membrane (approximately 3 h per sample), which is more time-consuming than the ultracentrifugation method (less than 2 h per six samples), the large-volume ultracentrifugation method was chosen for further comparison with the ultracentrifugation-based small-volume method which had the highest recovery efficiency among the eight small-volume methods.

Similar Ct values were obtained for the small-volume and large-volume ultracentrifugation methods (Figure 3b ). For viral concentrations at a marginal level of 1 -100 copies/mL wastewater, both the large-and small-volume methods reached the same J o u r n a l P r e -p r o o f virus concentration range of 10 copies/mL wastewater (medium concentration) (Table 3 ).

In addition, the detection rates of the small-and large-volume methods were the same for 35 wastewater samples without spiking (Table 3) . Only slightly lower Ct values were observed in the large-volume method than in the small-volume method with a Ct difference of 0.67 (Figure 3b) . The results show that increasing the sample volume did not significantly increase the detection sensitivity, probably because the wastewater matrix was co-enriched with viral particles using the large-volume method, unfavorable for the viral RNA extraction and detection. This implies the need to quantify the effects of the wastewater matrix on viral RNA extraction and detection for a more comprehensive understanding of the quantification of SARS-CoV-2 in wastewater samples.

Two viral RNA extraction methods were further compared using SARS-CoV-2 spiked wastewater samples and 31 unspiked wastewater samples using the small-volume ultracentrifugation method. Lower Ct values were observed using the lysis-buffer-based method (Viral Kit) than the acid-guanidinium-phenol-based method (TRIzol Kit) for spiked wastewater from different sampling sites, which represented varied matrix compositions (Figure 4a ). In addition, for spiked wastewater at marginal levels, the Viral Kit can detect SARS-CoV-2 RNA at "Low" concentration (1 copy/mL wastewater), representing good sensitivity using Viral Kit (Table 3 ). In contrast, TRIzol Kit could only detect RNA low to "Medium" concentration (10 copies/mL wastewater) level. Finally, the Viral Kit yielded higher detection rates and lower Ct values than the TRIzol Kit in 31 J o u r n a l P r e -p r o o f unspiked wastewater samples (Table 3 , Figure 4b ). These results indicate that the Viral Kit may perform better while both of the two kits had good performance.

The differences in detection rates and Ct values may be due to higher virus extraction efficiency in the Viral Kit rather than in the TRIzol Kit, especially when processing relatively particle-free samples. To confirm this hypothesis, serial dilutions of SARS-CoV-2 virus in PBS or wastewater were extracted using two viral extraction kits. As shown in Figure 4c and Figure 4d , the Viral Kit consistently obtained lower Ct values than the TRIzol Kit for a dynamic range of 3 -8 log 10 copies/reaction, irrespective of the sample type. The Ct values differences ranged 2.86 -4.18 and 0.86 -2.99 for PBS solution and wastewater samples, respectively. These results suggest the Viral Kit obtain more viral RNA for detection than the TRIzol Kit in the same samples.

Compared with the TRIzol Kit, the Viral Kit is more rapid in experimental procedures and more commonly used in SARS-CoV-2 WBE studies (Haramoto et al., 2020; Hata et al., 2021) . The Viral Kit is applicable to automated extraction machines such as QIAcube Connect (Goncalves et al., 2021) , permitting high-throughput viral RNA extraction in the rapid implementation of wastewater surveillance during a pandemic. However, the Viral Kit has a disadvantage of spin column clogging issues, whereas the TRIzol Kit present superior performance in turbid samples, such as large-volume wastewater samples in this study, and wastewater samples concentrated by PEG precipitation (Torii et al., 2021) .

In the present study, ultracentrifugation outperformed other concentration methods when processing 30 mL of wastewater, but we cannot overlook the sensitivity increase J o u r n a l P r e -p r o o f potential of other concentration methods when processing larger sample volumes. In addition, low-speed centrifugation (4750 × g for 30 min) was conducted before the virus concentration, which could enhance method sensitivity by removing debris but it may also influence the performance of precipitation-based methods evaluated in this study.

Meanwhile, different recovery efficiencies of small-volume ultracentrifugation methods were observed among different sampling sites due to matrix effects imposed by the wastewater samples. The performance of different virus concentration methods will be influenced by their tolerance to the matrix effect of the wastewater samples, which needs to be re-considered and re-evaluated when selecting a suitable concentration method for wastewater samples from other sampling sites in different countries.

Furthermore, the sensitivity of other virus concentration methods could be increased by modifying their method operational parameters or combining them with other virus RNA extraction kits. Therefore, future studies should explore the optimal operational parameters for a specific method and the selection of virus RNA extraction kits.

Finally, in addition to the method sensitivity evaluated in the present study using Hong Kong wastewater, the selection of virus concentration methods relied on equipment accessibility, method familiarity, labor requirement, supply availability, and throughput of the sample processing in the laboratory.

Using heat-inactivated SARS-CoV-2 virus spiked wastewater and unspiked wastewater samples, the present study assessed the performance of eleven virus concentration methods for detecting SARS-CoV-2 virus. The comparison of eight small-volume (30 J o u r n a l P r e -p r o o f mL) methods showed that ultracentrifugation was the most sensitive method, with the lowest Ct values and highest recovery efficiency (25.4 ± 5.9%). To explore the potential to increase method sensitivity, larger-volume concentration methods were compared with the small-volume method using ultracentrifugation. The sensitivity of the evaluated largevolume methods was similar to that of the small-volume method regarding detection rates and Ct values. Considering operational feasibility and throughput, processing smallvolume wastewater samples is a cost-effective, sensitive, and robust strategy.

Furthermore, using the small-volume ultracentrifugation method, two viral RNA extraction methods, i.e. lysis-buffer-based method vs acid-guanidinium-phenol-based method, were compared, and the results showed that the lysis-buffer-based was more sensitive, with a higher processing capacity and less time demand than the acidguanidinium-phenol-based method. Overall, the present study found that the combination of ultracentrifugation and lysis buffer-based RNA extraction method could be used for the rapid and sensitive detection of SARS-CoV-2 in the wastewater surveillance. The findings of our study provide recommendations and reference for the future applications of existing and newly developed virus concentration methods for SARS-CoV-2 wastewater surveillance.

J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f  Blank: raw wastewater without spiking SARS-CoV-2 virus;

 Low: SARS-CoV-2 virus-spiked wastewater with a concentration of approximately 1 copy/mL wastewater;

 Medium: SARS-CoV-2 virus-spiked wastewater with a concentration of approximately 10 copies/mL wastewater;

 High: SARS-CoV-2 virus-spiked wastewater with a concentration of approximately 100 copies/mL wastewater;

 Total detection rates: the percentage of detectable samples for the chosen method;

 Single detection rates: the percentage of only detectable samples for the chosen method but not detectable using the comparable method;

 Lower Ct: the percentage of showing lower Ct values for the chosen method when detectable using both methods.

J o u r n a l P r e -p r o o f

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

Assessment of inactivation procedures for SARS-CoV-2

Heat inactivation of the severe acute respiratory syndrome coronavirus 2

Wastewater Surveillance during Mass COVID-19 Vaccination on a College Campus

Epidemiology of the silent polio outbreak in Rahat, Israel, based on modeling of environmental surveillance data

Methods for primary concentration of viruses from water samples: a review and meta-analysis of recent studies

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

Stability of SARS-CoV-2 in different environmental conditions

Molecular Diagnosis of a Novel Coronavirus (2019-nCoV) Causing an Outbreak of Pneumonia

Early-pandemic wastewater surveillance of SARS-CoV-2 in Southern Nevada: Methodology, occurrence, and incidence/prevalence considerations

Implementing building-level SARS-CoV-2 wastewater surveillance on a university campus

Detection of SARS-CoV-2 RNA in hospital wastewater from a low COVID-19 disease prevalence area

Near realtime determination of B.1.1.7 in proportion to total SARS-CoV-2 viral load in wastewater using an allele-specific primer extension PCR strategy

Quantification of SARS-CoV-2 and cross-assembly phage (crAssphage) from wastewater to monitor coronavirus transmission within communities

Potential intestinal infection and faecal-oral transmission of SARS-CoV-2

First environmental surveillance for the presence of SARS-CoV-2 RNA in wastewater and river water in Japan

Detection of SARS-CoV-2 in wastewater in Japan during a COVID-19 outbreak

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

Monitoring SARS-CoV-2 in municipal wastewater to evaluate the success of lockdown measures for controlling COVID-19 in the UK

The devil is in the details: emerging insights on the relevance of wastewater surveillance for SARS-CoV-2 to public health

Benchmarking virus concentration methods for quantification of SARS-CoV-2 in raw wastewater

Detection of SARS-CoV-2 variants in Switzerland by genomic analysis of wastewater samples

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

How sewage could reveal true scale of coronavirus outbreak

Development of a large volume concentration method for recovery of coronavirus from wastewater

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

Prevalence in the Early Stage of the Epidemic in The Netherlands

Testing at scale during the COVID-19 pandemic

Temporal Detection and Phylogenetic Assessment of SARS-CoV-2 in Municipal Wastewater

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

Comparing analytical methods to detect SARS-CoV-2 in wastewater

A comparison of SARS-CoV-2 wastewater concentration methods for environmental surveillance

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

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

Applicability of polyethylene glycol precipitation followed by acid guanidinium thiocyanate-phenol-chloroform extraction for the detection of SARS-CoV-2 RNA from municipal wastewater

Virological assessment of hospitalized patients with COVID-2019

SARS-CoV-2 Titers in Wastewater Are Higher than Expected from Clinically Confirmed Cases. mSystems 2020

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

The first case study of wastewater-based epidemiology of COVID-19 in Hong Kong

Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China

Figure 2 Comparison of eight small-volume concentration methods. (a) Ct values; (b) recovery efficiency (%)

MgCl 2 precipitation (1%, 2.5M, v/v); A15: using a Amicon-Ultra 15 Centrifugal Filter with a cut-off of 10 kDa; C70: using a Centricon Plus-70 centrifugal filter with a cut-off of 30 kDa

MgM: using a 0.45 μm electronegative membrane with the addition of MgCl 2 solution (1%, 2.5M, v/v)

Experiments were repeated in duplicate (n = 2). (b) Ct values of ultracentrifugation methods based on 31 unspiked wastewater samples (n = 31). (c) Ct values of PBS serial dilution SARS-CoV-2 virus; Experiments were repeated in triplicate (n = 3). (d) Ct values of wastewater serial dilution SARS-CoV-2 virus

Resources. Chung In Yau: Methodology. Alex W.H. Chin: Resources. Leo L.M. Poon: Methodology, Resources, Writing -review & editing. Hein Min Tun: Resources, Writing -review & editing. Tong Zhang: Conceptualization

This study was financially supported by Health and Medical Research Fund (HMRF) (COVID190209 and COVID190116), the Food and Health Bureau, The Government of

MS2 mean 1.0% -9.5% (Torii et al., 2021) φ6 mean 1.6% -9.7% (Torii et al., 2021) J o u r n a l P r e -p r o o f