High-throughput sequencing of SARS-CoV-2 in wastewater provides insights into circulating variants

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged from a zoonotic spill-over event and has led to a global pandemic. The public health response has been predominantly informed by surveillance of symptomatic individuals and contact tracing, with quarantine, and other preventive measures have then been applied to mitigate further spread. Non-traditional methods of surveillance such as genomic epidemiology and wastewater-based epidemiology (WBE) have also been leveraged during this pandemic. Genomic epidemiology uses high-throughput sequencing of SARS-CoV-2 genomes to inform local and international transmission events, as well as the diversity of circulating variants. WBE uses wastewater to analyse community spread, as it is known that SARS-CoV-2 is shed through bodily excretions. Since both symptomatic and asymptomatic individuals contribute to wastewater inputs, we hypothesized that the resultant pooled sample of population-wide excreta can provide a more comprehensive picture of SARS-CoV-2 genomic diversity circulating in a community than clinical testing and sequencing alone. In this study, we analysed 91 wastewater samples from 11 states in the USA, where the majority of samples represent Maricopa County, Arizona (USA). With the objective of assessing the viral diversity at a population scale, we undertook a single-nucleotide variant (SNV) analysis on data from 52 samples with >90% SARS-CoV-2 genome coverage of sequence reads, and compared these SNVs with those detected in genomes sequenced from clinical patients. We identified 7973 SNVs, of which 5680 were novel SNVs that had not yet been identified in the global clinical-derived data as of 17th June 2020 (the day after our last wastewater sampling date). However, between 17th of June 2020 and 20th November 2020, almost half of the SNVs have since been detected in clinical-derived data. Using the combination of SNVs present in each sample, we identified the more probable lineages present in that sample and compared them to lineages observed in North America prior to our sampling dates. The wastewater-derived SARS-CoV-2 sequence data indicates there were more lineages circulating across the sampled communities than represented in the clinical-derived data. Principal coordinate analyses identified patterns in population structure based on genetic variation within the sequenced samples, with clear trends associated with increased diversity likely due to a higher number of infected individuals relative to the sampling dates. We demonstrate that genetic correlation analysis combined with SNVs analysis using wastewater sampling can provide a comprehensive snapshot of the SARS-CoV-2 genetic population structure circulating within a community, which might not be observed if relying solely on clinical cases.

Figure 1:

A. Map of the United States of America with states where wastewater samples were collected for this study highlighted in grey. B. SARS-CoV-2 RT-qPCR Ct detection value for each sample and the corresponding SARS-CoV-2 genome coverage uniformity from the tiling amplicon-based HTS. C. SARS-CoV-2 genome coverage of the high-throughput sequencing of all the wastewater samples (cyan) and those with >90% coverage (red). * indicates that these sites have a coverage depth of 1.

Figure 2:

A. Number of single nucleotide variants (SNV) per sample across 10 states (each state is represented by a different colour). B. Regression analysis, with 95% confidence interval, of the number of wastewater-derived SARS-CoV-2 SNVs detected versus the mean depth for each of the 52 samples with >90% coverage that were analysed. The colour code indicates the states in which the samples were collected.

Figure 3:

Novel SARS-CoV-2 SNVs (i.e. not yet detected in clinical-derived samples as of 17^th June 2020) identified in the 52 wastewater samples analysed. On the y-axis are the number of samples containing the SNV and on the x-axis is the relative position of SNV in the SARS-CoV-2 genome. Positions with multiple variants are marked in red and those marked with grey circles represent the SNVs that have been detected up until 20^th November 2020 in clinical samples.

Figure 4:

Publicly available genomes from clinically derived data deposited in GISAID, grouped by PANGOLIN, whose mutations were consistent with those observed in wastewater samples. A. Heatmap showing the number of days between sample collection and when supported lineages were first observed in clinical data. Each wastewater sample (52 samples across 10 states) contained support for different clinical samples which are grouped here by PANGOLIN, some of which have only been observed outside North America (indicated as “global only”). B. Clinical genomes reported in USA states and territories which were assigned to PANGOLIN supported by at least one environmental sample. Black borders indicate lineages supported in environmental samples from the respective location.

Figure 5:

Principal coordinate analysis (PCoA) of SARS-CoV-2 sequence data derived from wastewater samples. A. Distribution of sequences from samples collected in ten states (each represented by a different colour) in the USA showing pairwise distance based on genomic composition between viral populations present in each sample. B. Timeline representation (shown by the colour gradient) of samples taken from the sample locations across ten USA states between April-June 2020 with pairwise distance based on genomic composition between viral populations present in each sample. C. Spatial representation of SARS-CoV-2 sequences from samples collected from various regions within Arizona (represented by different symbols) comparative to those from other states. D. Sampling catchments in Tempe, Guadalupe and Gilbert, Arizona.