Detecting COVID-19 variants in wastewater


The global COVID-19 pandemic escalated quickly, and the challenge has been compounded by the rapid spread of new variants. The Source reviews some of the ongoing efforts of the water science community to apply wastewater surveillance to these variants.

“Every country in the world is struggling to control the spread of variants,” says Professor Banu Ormeci, co-chair of the IWA COVID-19 Task Force, summing up the current situation in a recent IWA webinar convened by the Task Force.

The emergence of variants has sparked major concerns regarding increased transmissibility and infectivity, disease severity, and potential resistance to vaccines and treatments.

A variant, now designated the Alpha variant by WHO, was detected in the UK towards the end of 2020 and, despite a harsh lockdown, the Dutch authorities confirmed that, in just 12 weeks, the variant had not only entered the Netherlands, but had come to represent almost 90% of cases. This illustrates how the global situation escalated quickly, both in terms of the spread of the disease and the emergence of variants.

Alongside this, the water science community responded at speed. To gather early experiences, the IWA COVID-19 Task Force brought together leading experts in a webinar to share their experiences of using a range of methods to identify and track SARS-CoV-2 Variants of Concern (VoC) in their countries.

Sewage surveillance

Monitoring wastewater for biological content such as pathogens and viruses has long been a regular function of water utilities. The pandemic, however, has catapulted the role of sewage surveillance as a wider public health ‘early warning tool’.

Researchers, including Professor Gertjan Medema, principal microbiologist at KWR, identified early on in the pandemic that wastewater surveillance could be used to detect the presence of SARS-CoV-2 at the community level, identifying trends and outbreaks. Since then, the rapid emergence of variants has intensified interest in its use to contribute to tracking, and containing, specific ‘variant’ outbreaks that are distinguishable from a dominant circulating strain, referred to as the ‘wild type’.

Searching for variants is now a specialism within the area of wastewater coronavirus monitoring. While the science involved in identifying VoCs is complex, it does not detract from the regular surveillance of SARS-CoV-2.

Medema, along with Dr Tamar Kohn, of the Swiss Federal Institute of Technology, and Giuseppina La Rosa, from the Italian National boldenone undecylenate bijwerkingen Institute of Health, provided latest updates to the audience of their evolving methods to identify VoCs in municipal wastewater. The role of variant surveillance is to observe the presence and dynamics of the virus and VoCs, including potential impacts of VoCs on vaccination efficacy. Identifying the emergence of VoCs and tracking their presence spatially and over time provides vital information to governing and medical authorities to support the establishment of targeted preventative interventions.

“Identifying the emergence of VoCs and tracking their presence spatially and over time provides vital information to governing and medical authorities”

Shared concepts: genome sequencing and PCR

Two fundamental approaches underpin current VoC surveillance methods: virus genome ‘sequencing’ routines, and polymerase chain reaction (PCR)-based techniques. The PCR-based techniques, which identify the presence and quantity of SARS-CoV-2 RNA, can also detect signature mutations, which are characteristic of variants. Each mutation is allocated a unique code, and variants typically have a combination of mutations. For example, the Alpha (‘UK’) VoC contains the N501Y, E484K and K417 mutations. The final of these three mutations differentiates the Alpha variant from the Beta (‘South African’) and Gamma (‘Brazilian’) variants, for which the mutation is designated K417N and K417T, respectively.

Both techniques are currently being tested by KWR in the Netherlands in ‘next generation sequencing’ (NGS), with bioinformatics to analyse the SARS-CoV-2 genome and digital droplet PCR (DD-PCR) analysis, which detects rare occurrences of mutant virus in a sea of ‘wild type’ SARS-CoV-2. DD-PCR basically looks for genome sequences in wastewater that are very similar to the sequences found in people living in the sewerage catchment.

“Studies so far indicate that the process can be used to test for VoCs using individual digital droplets, containing very small units of the virus sequence,” said Medema, adding that “it’s possible to identify the variant when it represents just 1% (or less) within a mixture of wild type.”

NGS using bioinformatics analysis is also being developed at the Swiss Federal Institute of Technology, in work led by Tamar Kohn. The Swiss method is highly targeted, analysing genetic variation in specific genomic regions of the virus, and sequencing PCR products (amplicons). “Viral RNA is extracted from raw wastewater after ultrafiltration, which removes a lot of PCR inhibitors, remains sufficiently intact and is well-suited for NGS analysis,” comments Kohn. The analytics use a bespoke bioinformatics platform, V-Pipe, developed by a computational biology team at ETH Zurich, led by Professor Niko Beerenwinkel. This ultimately enables an analytical process of ‘mutation calling’ on each amplicon.

Work led by Giuseppina La Rosa at the Italian National Institute of Health is developing an alternative approach, ‘nested RT-PCR (reverse transcriptase PCR) and Sanger sequencing’. Sanger sequencing was developed in 1977 by Frederick Sanger and targets specific fragments of DNA to access genomic information within nucleotides. It is the same process that was used to sequence the entire human genome.

Sequencing maps genetic code from the full SARS-CoV-2 genome and from individual ‘long PCR fragments’, enabling multiple mutations to be detected. However, amplifying the long fragments from the challenging wastewater matrix is far from simple, so researchers are currently testing the use of two short-fragment nested RT-PCRs to improve the method for wastewater.

“So far, all variants identified in clinical observations have been identified and correctly assigned within the code fragments, increasing confidence in the technique,” said La Rosa. The process detected the Alpha, Beta and Gamma variants, as well as those emerging from Spain, California and Denmark. Even the newest VoCs emerging in Nigeria, India and Japan were detectable in the same long fragment of the original spike protein. As with other studies, the approach being developed in Italy has been tested and validated against clinical samples before being applied to environmental samples.

The focus of each of these techniques is on detecting the emergence and prevalence of VoCs, rather than relating amounts of virus in sewage to the level of infection in the population. The specific questions that the analyses are used to answer are: whether the variant has entered the population; when it entered; how prevalent it is; and whether that prevalence is changing.

Variants rise to dominance

In the Netherlands, approximately 1,000 random clinical patient samples are collected per week from hospitals across the country to track coronavirus variants, providing a clinical context against which to compare the wastewater samples. It currently takes three to four weeks to process the clinical data.

NGS outputs show how the emergence and development of a variant can be identified and tracked geographically and over time through wastewater analyses. The figures show how quickly the Alpha variant became established in the Netherlands during January and February 2021, including in Amsterdam, becoming prevalent within two months. By the end of February, the Alpha variant had reached 85% in Amsterdam and Utrecht. By April 2021, it had become dominant, in effect becoming the new SARS-CoV-2 ‘wild type’ against which new variants will be monitored.

Interestingly, studies in the Netherlands have shown that wastewater sequences also reveal the presence of rare mutations not observed in patient data. It is possible to speculate that these mutations represent new variants within the population but, critically, wastewater is not being used to indicate which variants are of concern – that remains the responsibility of epidemiological studies.

In April 2021, the Alpha variant was the most abundant in Switzerland. The Swiss researchers have also focused their monitoring efforts on the Alpha variant, and, according to Dr Kohn, are “very excited about their results”, which show a similar emergence and rise to dominance trend as seen in the Netherlands.

The decision to open ski resorts during the pandemic was highly contentious, but a popular alpine resort has provided a valuable sample point for researchers. The sequencing data enabled scientists at the Swiss Federal Institute of Technology to reconstruct the emergence and prevalence timeline of the variant at three locations: the ski resort (representing a ‘pop-up’ sample zone during December 2020 with a population equivalent (PE) of just 10,000), and longer-term wastewater monitoring in Zurich (PE 450,000) and Lausanne (PE 240,000).

According to Kohn, the amplicon process “can detect combinations of mutations and is more powerful than analysing mutations individually, and provides very strong evidence for the presence of a VoC”. Amplicons containing the Alpha variant mutations were identified in Lausanne in early December, with a statistically significant signal matching the clinical data.

Trade-offs for rapid screening

The results of the Italian studies have been more mixed. In environmental samples, Sanger sequencing did not detect the Alpha variant in the Italian Umbrian region when occurrence levels were low, suggesting it may underestimate less prevalent strains. The trial, using two short-fragment PCR assays instead of one longer fragment, has shown that this approach may result in each test amplifying a different target. Inevitably, the quantity of information retained on shorter fragments is reduced, but results indicate that both methods can discriminate between VoCs, as well as other viruses. Further work is ongoing to optimise the long-fragment PCR method for environmental samples, and to combine it with NGS or long-read sequencing for more in-depth sequence analyses.

However, nested RT-PCR does provide a rapid screening method for VoCs in water treatment plant catchments. This could accelerate action in areas where clinical surveillance and/or targeted preventative intervention is required. It can be performed routinely, at low cost, and is relatively easy to interpret.

Wastewater versus clinical sequencing

Analysing wastewater for VoCs can be seen as being more complex than analysing clinical samples. Wastewater contains a mixture of variants from many people, so analyses need to be able to distinguish between the relevant combinations of mutations to understand which variants are observed and, therefore, present in the population. The research is relatively new, and testing and verifying the methods and results against clinical data remains a key metric of the validity of wastewater monitoring. Despite following different analytical methods, all of the studies so far have reported their findings in comparison with clinical data with high levels of confidence.

Looking ahead

According to a poll conducted during the IWA webinar, two thirds of country respondents confirmed they are considering using wastewater as a tool to monitor VoCs. Based on experience in the Netherlands, Professor Medema advises countries “to build consortia involving health institutions, research institutions, and water utilities, as, ultimately, the health institutions as the ‘end user’ are critical to turn wastewater research into public health action”.

Looking ahead, the intention is to detect variants before they appear in hospitals. Studies last year (August 2020) found testing at wastewater treatment plants was sensitive enough to identify virus circulation before the virus was observed in clinical reports. It is hoped that the same will be true for identifying VoCs.

The benefits of sequencing are becoming clear, but challenges persist that could inhibit sequencing for VoC monitoring, principally a lack of trained bioinformatics personnel and the high cost and lack of sequencing capacity. With the prospect of coronavirus continuing to plague the world for many years to come, and anticipating that we will face other similar threats in the future, it is clear there will be a need for more data and more bioinformatics specialists to develop this technology for the future. Meanwhile, with nested RT-PCR providing a rapid screening method for VoCs in catchments, there is the prospect of accelerated action in areas where clinical surveillance and/or targeted preventative intervention are needed. •

More information

Detecting COVID-19 variants in wastewater. IWA webinar, 13 April 2021. Watch the webinar at:

Rapid screening for SARS-CoV-2 Variants of Concern in clinical and environmental samples using nested RT-PCR assays targeting key mutations of the spike protein, G La Rosa, et al, Water Research.

COVID-19 Variants of Concern

WHO label                  Lineage                      Earliest documented samples

Alpha                           B.1.1.7                         United Kingdom, September 2020

Beta                             B.1.351                        South Africa, May 2020

Gamma                       P.1                               Brazil, November 2020

Delta                            B.1.617.2                     India, October 2020

Source: WHO, 31 May 2021.