Estimating SARS-CoV-2 Through Wastewater Surveillance

Methods

Mutation Assay Development for Alpha and Delta Variants

We developed assays in silico to target mutations present in Alpha (HV69–70) and Delta (Del156–157/R158G). We screened primers and probe sequences (Appendix Table, https://wwwnc.cdc.gov/EID/article/28/5/21-2488-App1.pdf) for specificity using BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi), and then tested them in vitro against a wide range of viral genomes, including wild-type SARS-CoV-2 and SARS-CoV-2 VOCs, including Alpha and Delta. We further tested the sensitivity and specificity of the assays by diluting variant gRNA containing the mutations in no (0 copies), low (100 copies), and high (10,000 copies) background of wild-type gRNA (Appendix).

Wastewater Sample Collection

This study used samples from 2 publicly owned treatment works (POTWs) that serve ≈1.5 million residents of Santa Clara County, California, USA (San Jose), and Sacramento County, California, USA (Sacramento). Details of collection processes have been described.^[14]

We collected samples from the POTWs to span the period before and including the presumed emergence of Alpha and Delta variants in the communities. Before presumed emergence, sampling was 1–4 times per month; during the periods of suspected emergence, sampling was 3–7 times per week. At San Jose, 133 (HV69–70) and 48 (del156–157/R158G ) samples and at Sacramento, 64 (HV69–70) and 48 (del156–157/R158G) samples were included for analyses of each mutation.

We extracted RNA from the settled solids and processed within 24 hours of sample collection to measure concentrations of the nucleoprotein (N) gene using digital droplet reverse transcription PCR (Appendix).^[20] The N gene codes for the SARS-CoV-2 nucleocapsid; the specific region of the genome targeted by the assay is conserved on SARS-CoV-2 genomes. We included internal recovery controls. Thereafter, we stored RNA samples at –80°C for 0–300 days before analyzing them a second time for the N gene and the Delta mutation (Del156–157/R158G) or the Alpha mutation (HV69–70), using digital droplet reverse transcription PCR. By comparing the N gene concentration in the samples before and after storage, we confirmed negligible RNA degradation. All wastewater data are publicly available (https://doi.org/10.25740/zf117dn1545).

Incident COVID-19 Cases and Case Isolate Sequences

Each POTW provided sewershed boundary shapefiles. We determined the number of PCR-confirmed COVID-19 cases reported to CDPH as a function of episode date (earliest of either specimen collection or symptom onset date) residing within each sewershed using methods reported previously^[20] (Appendix).

COVID-19 case isolate whole-genome sequence data available to CDPH included data from the CDC and laboratory partners. We assigned sequence data to a sewershed on the basis of residential home postal code for the sample. We assigned the PANGO lineage based on the software version available at the time data was extracted; most recent results used pangoLEARN and pango-designation version 1.2.66.^[21]

We calculated VOC abundance estimates by dividing the number of sequences identified as Alpha or Delta (using the World Health Organization definition and including all PANGO sublineages Q.*, for Alpha, and AY.*, for Delta) by the total number of isolates sequenced from persons residing in the sewersheds over 14-day periods. To estimate time between isolate sample collection and sequence result and to measure the effect of that time delay on VOC estimates, we compared 14-day VOC abundance estimates over time against a final estimate generated on August 24, 2021. We chose a 14-day VOC window (versus a 7- or 28-day window) to balance timeliness of results and number of available case isolates sequenced within the window, given that fewer case isolates increase the uncertainty of estimates.

We performed Pearson correlations between the wastewater mutation and case isolate variant datasets, comparing the mean ratio of mutations in wastewater (HV69–70 and Del156–157/R158G to the N gene) to the proportion of case isolates sequenced and characterized as Alpha or Delta, each averaged over the previous 14 days. We used 0 as a replacement for samples where the measurement was below the limit of detection (nondetect); we repeated the analysis by using half the detection limit (500 copies/g), and the results were the same. We set statistical significance at p<0.05 and performed analyses in R studio version 1.4.1106 (https://www.rstudio.com).