Assessing Multiplex Tiling PCR Sequencing Approaches for Detecting Genomic Variants of SARS-CoV-2 in Municipal Wastewater

Abstract

Wastewater-based genomic surveillance of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus shows promise to complement genomic epidemiology efforts. Multiplex tiling PCR is a desirable approach for targeted genome sequencing of SARS-CoV-2 in wastewater due to its low cost and rapid turnaround time. However, it is not clear how different multiplex tiling PCR primer schemes or wastewater sample matrices impact the resulting SARS-CoV-2 genome coverage. The objective of this work was to assess the performance of three different multiplex primer schemes, consisting of 150-bp, 400-bp, and 1,200-bp amplicons, as well as two wastewater sample matrices, influent wastewater and primary sludge, for targeted genome sequencing of SARS-CoV-2. Wastewater samples were collected weekly from five municipal wastewater treatment plants (WWTPs) in the Metro Vancouver region of British Columbia, Canada during a period of increased coronavirus disease 19 (COVID-19) case counts from February to April 2021. RNA extracted from clarified influent wastewater provided significantly higher genome coverage (breadth and median depth) than primary sludge samples across all primer schemes. Shorter amplicons appeared to be more resilient to sample RNA degradation but were hindered by greater primer pool complexity in the 150-bp scheme. The identified optimal primer scheme (400 bp) and sample matrix (influent) were capable of detecting the emergence of mutations associated with genomic variants of concern, for which the daily wastewater load significantly correlated with clinical case counts. Taken together, these results provide guidance on best practices for implementing wastewater-based genomic surveillance and demonstrate its ability to inform epidemiology efforts by detecting genomic variants of concern circulating within a geographic region. IMPORTANCE Monitoring the genomic characteristics of the SARS-CoV-2 virus circulating in a population can shed important insights into epidemiological aspects of the COVID-19 outbreak. Sequencing every clinical patient sample in a highly populous area is a difficult feat, and thus sequencing SARS-CoV-2 RNA in municipal wastewater offers great promise to augment genomic surveillance by characterizing a pooled population sample matrix, particularly during an escalating outbreak. Here, we assess different approaches and sample matrices for rapid targeted genome sequencing of SARS-CoV-2 in municipal wastewater. We demonstrate that the optimal approach is capable of detecting the emergence of SARS-CoV-2 genomic variants of concern, with strong correlations to clinical case data in the province of British Columbia. These results provide guidance on best practices on, as well as further support for, the application of wastewater genomic surveillance as a tool to augment current genomic epidemiology efforts.

Keywords: COVID-19; RNA; SARS-CoV-2; epidemiology; variants; wastewater; whole-genome sequencing.

FIG 1

SARS-CoV-2 whole-genome sequencing coverage results from three multiplex tiling PCR primer schemes, including breadth of genome coverage for 150-bp amplicons (A), 400-bp amplicons (B), and 1,200-bp amplicons (C), as well as the median depth of coverage across the genome for 150-bp amplicons (D), 400-bp amplicons (E), and 1,200-bp amplicons (F). The breadth of coverage represents the proportion of nucleotides in the SARS-CoV-2 reference genome (NCBI accession number MN908947.3) that are covered by at least one read, and the median depth of coverage represents the median number of reads mapped at each nucleotide position across the SARS-CoV-2 reference genome. Values are plotted versus the sample cycle threshold (C_T) value for the U.S. CDC N1 assay, measured by reverse transcription-quantitative PCR (RT-qPCR) (see Text S1 in the supplemental material). Data points aligned with the x axis (plots D to F) had values of zero and could not be log transformed.

FIG 2

(A) Frequency of single-nucleotide variants (SNVs) associated with the P.1 lineage of SARS-CoV-2 within influent wastewater samples from five wastewater treatment plants in Vancouver, British Columbia (BC), from 7 February to 18 April 2021. Smaller gray dots represent the frequency of individual variant of concern (VoC)-associated SNVs on the sample dates, while the larger black points represent the mean across all detected VoC-associated SNVs. Only genome positions with a read coverage over 50 are included in SNV frequency calculations. The VoC-associated SNVs are described in Text S1 in the supplemental material and provided in Table S3 at https://doi.org/10.6084/m9.figshare.16416528. (B) Frequency of the P.1 lineage in clinical COVID-19 patient cases in the province of BC, Canada, over the study period. The frequencies in clinical patient cases correspond to average values detected over an epidemiology (epi) week and were adapted from reference . (C) Correlation between the wastewater cumulative daily load of P.1 genomes summed across all five wastewater treatment plants (WWTPs) and the total P.1 clinical cases in the province of BC observed within the same epidemiological week. The wastewater P.1 daily load (genome copies/day) was approximated by normalizing copies of the SARS-CoV-2 N1 gene (copies/liter) by daily flow rates (liters/day) to obtain N1 loads (copies/day) for all WWTPs and multiplying those by the mean frequency of P.1-associated SNVs in each WWTP across all sample dates. For each date, the cumulative P.1 daily load was determined by summing the P.1 loads across all five WWTPs. The P.1 clinical case counts by week were estimated from reference by multiplying total provincial COVID-19 case counts by the frequency of P.1 in clinical provincial cases.