Evaluating storage conditions and enhancement strategies on viral biomarker recovery for WBE applications

Since its first postulation in the late 1990s by Daughton and Ternes for the monitoring of chemical pollutants, the research field surrounding wastewater-based epidemiology (WBE) has gained remarkable momentum, evolving into a disease surveillance tool (Daughton & Ternes 1999; Mackuľak et al. 2021). More recently, during the COVID-19 pandemic, WBE was adapted to monitor infection trends to help ease the strain placed on the global healthcare systems by the alarming number of coronavirus infections (Pillay et al. 2022). While considerable progress has been made in the area of WBE, there is still a lack of comprehensive knowledge surrounding sample storage and its effect on the data generated for WBE (Mauro et al. 2020; Zahedi et al. 2021). While temperature is a crucial factor driving microbial degradation during storage, there may be several other variables that influence microbial behaviour and the stability of biomarkers in wastewater. The majority of the research conducted thus far focused on the degradation rates of SARS-CoV-2 biomarkers at temperatures above 4 °C to understand the survival and persistence under environmental conditions (Gundy et al. 2009; Ahmed et al. 2020; Bivins et al. 2020). A few studies have evaluated the impact of storage temperatures on SARS-CoV-2, but the results of these studies have shown substantial inconsistency. The study by Weidhaas et al. (2021) showed a reduction in SARS-CoV-2 biomarkers after incubation at −80 °C for a week, whereas the study by Hokajärvi et al. (2021) found no degradation at −20 or −75 °C after 84 days.

Considering the above, it is important to understand that wastewater is a complex environment containing numerous biological and chemical constituents that impact the physicochemical characteristics of the matrix. In addition to temperature, these properties may further impact viral persistence within the matrix (Amoah et al. 2022). There is evidence that suggests factors such as pH and total solids may also impact viral persistence in a sample matrix (Amoah et al. 2020, 2022; Bhatt et al. 2020; Forés et al. 2021; Hamouda et al. 2021). A comparative study by Oliveira et al. (2021) looked at viral persistence in several matrices including river water, tap water, unfiltered wastewater, and filtered wastewater. The results from these studies show that viruses may persist for longer periods of time in less complex matrices (in this case river water) (Bertrand et al. 2012; Oliveira et al. 2021).

Additionally, Hokajärvi et al. (2021) reported high concentrations of viruses within the solid matter present in wastewater. This necessitates the addition of a pretreatment step to dislodge these particles. Additionally, studies have observed increased adsorption of enveloped viruses to wastewater solids when compared to non-enveloped viruses owing to their structural differences (Ye et al. 2016). Previous studies have successfully applied ultrasonication and sodium pyrophosphate for the dislodgement of viral particles from solids (Danovaro et al. 2001; Brown et al. 2015). However, this has not been applied to enhance the recovery of SARS-CoV-2 and influenza A particles from wastewater samples. Herein this study focuses on investigating the combined effects of storage parameters (i.e. temperature and time) and the physicochemical characteristics of wastewater (pH, dissolved oxygen (DO), electrical conductivity (EC), total solids, total fixed solids, and chemical oxygen demand (COD)) on the degradation patterns of SARS-CoV-2 and influenza A. Additionally, different pretreatment methods (ultrasonication and sodium pyrophosphate) for the enhancement of viral recovery from wastewater solids are explored to improve the efficiency of WBE as a disease surveillance system.

A composite raw influent sample of 8 L was collected from a municipal wastewater treatment plant located north of Durban, South Africa, and transported on ice to the laboratory. The physicochemical characteristics of the sample, i.e. pH, temperature, DO, salinity, EC, and total dissolved solids (TDS) was done using a YSI 556 MPS Handheld Multiparameter Instrument (YSI) upon arrival to the laboratory. The COD was determined using the HACH high range (20–1,500 mg/L) COD test kits and total solids and total fixed solids was determined as described in Standard Methods (Baird et al., 2017). Samples were subsequently divided into two batches: the first batch was seeded with attenuated forms of SARS-CoV-2 strain USA/WA1/2020, and influenza A strain subtype H1N1 (Microbiologics, USA), of predetermined concentration of 10 and 2 copies/μl, respectively, and the second batch was unseeded wastewater influent sample. Both batches were homogenized using a magnetic stirrer at maximum speed for 10 min. The seeded samples acted as a reference point for monitoring viral degradation and allowed for the comparison against naturally occurring viral degradation in the wastewater samples. Samples were then divided into 500 ml aliquots and stored in 1 L polypropylene sampling bottles at 4, −20, −80 °C, and at room temperature (±25 °C) for a period of 84 days. Prior to analysis, the stored samples were thawed overnight at 4 °C and allowed to adjust to ±25 °C. Thereafter, the physicochemical characteristics of the stored sample was assessed. The unseeded wastewater influent samples were analysed every 14 days throughout the 84-day period, and the seeded samples were analysed on days 0, 42, and 84. All experiments and analysis were conducted in triplicate.

Viral particles were concentrated using polyethylene glycol (PEG) 6000 together with NaCl according to the methods described by Sapula et al. (2021). Briefly, 100 ml of samples (two batches of 50 ml) were clarified by an initial primary centrifugation at 5,000 × g for 30 min at 4 °C to remove large particles, and the resulting pellet was used for RNA extraction. To maximize viral recovery, the supernatant from the first round (primary) centrifugation was further precipitated overnight at 4 °C with 15% PEG 6000 and 2% NaCl followed by centrifugation at 12,000 × g for 90 min at 4 °C (Sapula et al. 2021). An additional centrifugation step at 5,000 × g for 5 min at 4 °C was conducted in a swinging bucket rotor to ensure any residual particles that may have attached to the tube was dislodged. The final combined pellets were re-suspended in 140 μl of 1X phosphate-buffered saline (PBS) for RNA extraction.

RNA was extracted from the pellets using the Qiagen QiAmp Viral RNA MiniKit (Hilden, Germany) following the manufacturer's protocols. Extracted RNA was eluted in 80 μl nuclease-free water and stored at −20 °C.

The purity and yield of the extracted RNA was assessed using the Implen Nanophotometer. Thereafter, the copy numbers of SARS-CoV-2 and influenza A was assessed by ddPCR using the One-Step RT-ddPCR Advanced Kit for Probes (Biorad, USA). The primer and probe sequences for each gene target can be found in Supplementary Appendix Table A1. Each 22 μl ddPCR reaction contained 5 μl supermix, 2 μl reverse transcriptase, 1 μl dithiothreitol, 1.98 μl each of the forward and reverse primers (10 μM), 0.55 μl of 10 μM probe, 4.49 μl nuclease-free water, and 5 μl template RNA (1 ng). The droplets were generated using the QXD Automated Droplet Generator, read using the QX200 Droplet Reader together with the QuantaSoft 1.7 software, and analysed using the QuantaSoft Analysis Pro 1.0 software (Biorad, USA). The limit of detection for the N2 (SARS-CoV-2) and INFA (Influenza A) genes were 0.2 and 0.18 copies/μl.

Mechanical (ultrasonication) and chemical (sodium pyrophosphate) pretreatment methods were evaluated for its efficiency in enhancing viral particle recovery from wastewater solids. The pellets obtained after primary centrifugation were weighed out (150 mg wet weight) and resuspended in 100 μl nuclease-free water and used for each pretreatment. The samples were sonicated at 4, 8, 12, and 16 kHz using a 125 W sonicating probe for 1 and 3 min on ice and was interrupted after every minute to prevent overheating (Danovaro et al. 2001). To test the effects of chemical pretreatment, sodium pyrophosphate, an ionic dispersant, was added to the sample to a final concentration of 10 mM and was incubated at room temperature for 15, 30, 45 min, and 1 h. The samples were centrifuged at 8,000 × g for 1 min at 4 °C from which 140 μl of the supernatant was removed for RNA extraction. The remaining pellet was resuspended in 140 μl of 1X PBS and used for RNA extraction separately as previously described. Following this, viral particles were quantified using ddPCR.

Microsoft Excel was used for data entry and for determining mean values and standard deviations. Additionally, RNA copy numbers obtained from ddPCR was converted to log₁₀ values prior to conducting any statistical analysis. The coefficient of correlation between storage temperature over a period of 84 days against physicochemical characteristics as well as viral RNA copy numbers and the physicochemical characteristics was determined by the Pearson's correlation test on R Studio with a significant correlation supported by a p-value of <0.05. A correlation coefficient was used to determine positive, negative, or no correlation. Any significant differences between the means of RNA concentration stored at each temperature was analysed by a one-way analysis of variance (ANOVA) followed by Tukey's post-hoc test using GraphPad Prism, version 10 (USA) with a significant correlation supported by a p-value of <0.05.

In order to effectively use WBE for disease surveillance, it is necessary to understand the dynamic nature of stored wastewater samples (Hokajärvi et al. 2021; Parra-Arroyo et al. 2023). This study aimed to investigate the combined effects of storage conditions and wastewater characteristics on viral degradation patterns, as well as enhancement strategies to improve viral recovery.

Visual inspection of the samples stored at 25 °C after 7 days revealed a distinctive change in colour from brown to black as well as the odour of hydrogen sulphide along with the formation of a distinct layer at the top of the sample. At this point, adhesion of particles (biofilm-like) to the polypropylene storage bottles was observed. A similar trend was observed for samples stored at 4 °C for 42 days. Since temperature is one of the most significant factors driving microbial processes (Kataki et al. 2021; Parra-Arroyo et al. 2023), the observation of septic sludge formation after storing wastewater influent at higher temperatures (i.e. 25 and 4 °C) is indicative of anaerobic microbial processes and chemical reactions, such as the reduction of sulphates to sulphides, as indicated by the smell of hydrogen sulphide (Muttamara 1996). In contrast, no discernible differences were observed in samples stored at lower temperatures (i.e. −20 and −80 °C).

Most studies only focus on the liquid fraction of the wastewater which could lead to discrepancies in the data (Hokajärvi et al. 2021). However, studies have reported on the presence of SARS-CoV-2 in the solids present in wastewater with some studies even showing a higher concentration of the virus in the settled solids when compared to influent wastewater (Graham et al. 2020; Hokajärvi et al. 2021; Kitamura et al. 2021; Simpson et al. 2021). For this reason, the supernatant and pellets were assessed separately to better understand viral particle interactions within the wastewater matrix.

Initial analysis of the influent samples revealed a higher concentration of SARS-CoV-2 RNA biomarkers in the supernatant (5.28 ± 0.14 log copies/100 ml) when compared to the pellet (4.96 ± 0.06 log copies/100 ml) (Supplementary Appendix Table A3), while influenza A had a higher concentration in the pellet (4.99 ± 0.12 log copies/100 ml) compared to the supernatant (4.87 ± 0.04 log copies/100 ml) (Supplementary Appendix Table A4). With the seeded samples, viral loads in the supernatant were higher when compared to the pellets for both SARS-CoV-2 (5.43 ± 0.01 log copies/100 ml)/(4.48 ± 0.01 log copies/100 ml) and influenza A (4.96 ± 0.03 log copies/100 ml)/(4.48 ± 0.04 log copies/100 ml).

A one-way ANOVA was conducted to determine the effect of storage temperature on SARS-CoV-2 degradation. A significant difference between SARS-CoV-2 degradation and at least three storage temperatures was observed (p < 0.0001) (Supplementary Appendix Table A5). Tukey's multiple comparisons test showed that SARS-CoV-2 degradation in the supernatants and pellets of the influent and seeded samples were significantly different between all storage temperatures (p < 0.0001).

A one-way ANOVA showed a significant difference between influenza A degradation and at least three storage temperatures (p < 0.0001) (Supplementary Appendix Table A6). Tukey's multiple comparisons test showed that influenza A degradation in the supernatants and pellets of the influent and seeded wastewater was significantly different between all storage temperatures (p < 0.0001).

SARS-CoV-2 and influenza A are enveloped viruses containing an outer protein layer that is susceptible to degradation in the external environment (Kataki et al. 2021). Exposure to high temperatures leads to the stimulation of extracellular enzymes capable of degrading viral particles or nucleic acid (Olson et al. 2004; John & Rose 2005; Gundy et al. 2009). Furthermore, the genomes of both viruses have a higher ratio of adenine and uracil bases (commonly known as the AT content) when compared to the ratio of guanine and cytosine, making them AT-rich (Boussier et al. 2020; Fumagalli et al. 2023). An AT-rich genome exhibits lower stability than a GC-rich genome (Chan et al. 2009). Thus, the structure and genome composition of SARS-CoV-2 and influenza A make them extremely susceptible to degradation accounting for the high degradation observed across all storage temperatures. However, while degradation was noted in samples stored at all temperatures, the difference in viral degradation at all temperatures were statistically significant. Additionally, in a wastewater sample, enveloped viruses are extremely susceptible to degradation due to the high concentration of disinfectants and other surfactants (Bernard et al. 2019). Similar studies by Weidhaas et al. (2021), Simpson et al. (2021), and Ahmed et al. (2020) all observed SARS-CoV-2 degradation in stored wastewater samples across the different storage temperatures. However, a study by Hokajärvi et al. (2021) found that SARS-CoV-2 remained stable for up to 84 days at 4, −20, and −75 °C. However, their experimental set-up differed from this study in that raw wastewater samples were not analysed between days 0 and 29 which may account for the differences in data. In the case of the samples stored at 4 °C, the high degradation in the supernatant and low degradation in the pellet can be associated with viral attachment to wastewater solids. While the degradation of SARS-CoV-2 differed from the study by Hokajärvi et al. (2021), attachment of viral particles to wastewater solids in stored wastewater influent was observed. However, the reason for attachment was not determined.

Certain changes, i.e. salinity, DO, pH, COD, total solids (TS) and total fixed solids (TFS), may have potentially impacted viral concentrations as determined by statistical correlations. Storage time showed significant negative correlations with SARS-CoV-2 and influenza A degradation in the supernatant and pellets of the influent stored at all temperatures, and the supernatants of the seeded wastewater stored at ±25, 4, and −20 °C. Changes in COD showed a significant positive correlation with viral degradation in the seeded wastewater supernatants at all temperatures. Changes in TS and TFS showed a significant positive correlation with viral degradation in wastewater influent supernatants and pellets at all temperatures. Changes in pH showed significant negative correlations with SARS-CoV-2 concentrations in seeded wastewater pellets stored at ±25 and −20 °C and significant positive correlations with SARS-CoV-2 concentrations in the wastewater influent pellet in samples stored at −80 °C and seeded wastewater pellets stored at 4 and −80 °C. Additionally, a significant positive correlation was observed between changes in pH and influenza A concentrations in seeded wastewater samples stored at all temperatures. Other significant correlations were observed and are shown in Figures 4 and 5.

Factors such as the pH of a matrix impact aggregation and adsorption properties of viruses by changing its surface charge (Mariani et al. 2018; Amoah et al. 2020; Corpuz et al. 2020; Hamouda et al. 2021). This is further supported by the observation of biofilms in the samples stored at 25 and 4 °C which may have impacted viral concentrations in the samples (Li et al. 2021) and by the lowest decrease of SARS-CoV-2 concentrations in the seeded wastewater pellets observed at 4 °C. This may not be apparent in the other samples due to low viral concentrations or due to viral structure which may impact viral stability and adhesion properties. Furthermore, based on the experimental set-up of this study, the ratio of whole viral particles to viral RNA may differ between the wastewater influent and the wastewater samples spiked with attenuated virus. As a result, the influent sample contained more viral genetic material than the seeded samples, contributing to the difference between their storage properties. While the structure of SARS-CoV-2 may differ from that of influenza A, the difference between the naturally occurring viruses and the spiked virus must also be accounted for. Influenza A contains a lipid envelope comprised of several surface glycoproteins which include hemagglutinin and neuraminidase surface glycoprotein, each having 16 and 9 known subtypes (Cheung & Poon 2007). Similarly, SARS-CoV-2 has a nucleocapsid protein and an envelope made up of a spike protein, membrane protein, and envelope protein (Wang et al. 2020). These proteins have a surface charge, and changes to this charge may influence viral aggregation and adsorption properties (Pielak & Chou 2011; Amoah et al. 2020). In addition to structural differences, the genomes of SARS-CoV-2 and influenza A have different polarities (Wang et al. 2020; Brázda et al. 2021; Piret & Boivin 2021; Szczesniak et al. 2023), which may have further influenced viral adsorption and aggregation.

Additional pretreatments such as chemical (sodium pyrophosphate) and physical (sonication) methods have been found to be beneficial in dislodging the attached viral particles from the sludge and improving the efficiency of the recovery process (Danovaro et al. 2001; Juel et al. 2021). In this study, the ionic dispersant, sodium pyrophosphate did not enhance viral recovery, while sonication did affect the concentrations of viruses within the sample matrix. Chemical treatment using sodium pyrophosphate did not enhance SARS-CoV-2 and influenza A recovery from fresh wastewater samples. However, sonication at frequencies ranging from 4 to 16 kHz resulted in an increased SAR-CoV-2 abundance in the supernatant. In the untreated sample, there was an initial concentration of 4.52 ± 0.18 log copies/100 ml SARS-CoV-2, after sonication there was an increase ranging from 3.30 to 35.65% as shown in Table 1. Sonication at 8 kHz for 3 min yielded the highest concentration increase of SARS-CoV-2. In contrast, influenza A had an initial concentration of 4.38 ± 0.04 log copies/100 ml of influenza A in the supernatant. After pretreatment, the highest increase observed was 0.28% after sonication for 1 min at 4 kHz. Thereafter, as the sonication frequency increased, influenza A copy numbers decreased. Previous studies by Wu & Liu (2009) reported damage to the viral particles/genetic material at higher sonicating frequencies, which was observed for influenza A. A study by Danovaro et al. (2001) found that sodium pyrophosphate had no impact while sonication at 47 kHz increased viral concentrations. In contrast, Brown et al. (2015) treatment using a 10 mM sodium pyrophosphate increased viral concentrations, while ultrasonication had no significant impact on viral concentrations.

It is evident from the increase in SARS-CoV-2 concentrations post-treatment that a substantial concentration of the virus is lost to the pellet leading to discrepancies in data. This necessitates the addition of a pretreatment step in the WBE workflow. However, an optimization step is crucial for wastewater samples due to the fact that sample matrix and the virus can differ considerably as seen in this study.

This study showed that storage temperature and time as well as changes to the physicochemical characteristics of the sample significantly impacted viral biomarker degradation. Based on the findings of this study, influenza A had the lowest degradation at −20 °C and SARS-CoV-2 at −80 °C. The differences in degradation between storage at −20 and −80 °C were significant.

It was further deduced that SARS-CoV-2 viral particles have an affinity for the solids in wastewater. This interaction between viral particles and wastewater solids may have been influenced by changes in the physicochemical characteristics of the sample. Based on the success of the pretreatment methods in dislodging viruses from the wastewater solids in fresh samples, this can also be applied to stored wastewater samples.

Recommendations made by standard microbiology guides suggest storage of environmental samples at 4 °C for 48 h and storage at −80 °C for prolonged periods (APHA 2002; Olson et al. 2004). It is evident that even at −80 °C, viral degradation is high, which raises concerns about the reliability of data obtained from stored samples. Degraded samples may produce misleading or inaccurate results, leading to erroneous conclusions. This necessitates the evaluation of wastewater preservation studies for WBE analysis for different pathogens.

Significant progress has been made in the field of WBE, supporting its use as a disease surveillance tool. However, as evidenced by the findings of this study, there are opportunities for refinement within the methodology. Based on this, the following recommendations can be made:

• Considering the high degradation even at recommended storage temperature, more emphasis needs to be placed on the importance of preservation techniques to maintain sample integrity for disease surveillance.

• Physicochemical characteristics were found to impact viral concentration, hence, further studies investigating the storage of post-centrifuged samples and concentrated samples should also be explored to potentially lessen the impact of sample matrix on the integrity of the viral particles.

• pH adjustments for long-term storage could potentially be explored to reduce viral affinity for wastewater solids and reduce microbial processes.

• Pre-treatment methods for the dislodgement of viral particles should be included in the WBE workflow for wastewater samples to enhance viral recovery in stored samples.

We are grateful for the support received from the Durban University of Technology and the financial support received from the National Research Foundation of South Africa.

This work was funded by the South African Research Chair Initiative (UID 84166) and the National Research Foundation of South Africa competitive funding for rated researchers (UID 129358).

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.