Comparing analytical methods to detect SARS-CoV-2 in wastewater

Wastewater based epidemiology (WBE) has emerged as a reliable strategy to assess the coronavirus disease 2019 (COVID-19) pandemic. Recent publications suggest that SARS-CoV-2 detection in wastewater is technically feasible; however, many different protocols are available and most of the methods applied have not been properly validated. To this end, different procedures to concentrate and extract inactivated SARS-CoV-2 and surrogates were initially evaluated. Urban wastewater seeded with gamma-irradiated SARS-CoV-2, porcine epidemic diarrhea virus (PEDV), and mengovirus (MgV) was used to test the concentration efficiency of an aluminum-based adsorption-precipitation method and a polyethylene glycol (PEG) precipitation protocol. Moreover, two different RNA extraction methods were compared in this study: a commercial manual spin column centrifugation kit and an automated protocol based on magnetic silica beads. Overall, the evaluated concentration methods did not impact the recovery of gamma-irradiated SARS-CoV-2 nor MgV, while extraction methods showed significant differences for PEDV. Mean recovery rates of 42.9 ± 9.5%, 27.5 ± 14.3% and 9.0 ± 2.2% were obtained for gamma-irradiated SARS-CoV-2, PEDV and MgV, respectively. Limits of detection (LoD(95%)) for five genomic SARS-CoV-2 targets (N1, N2, gene E, IP2 and IP4) ranged from 1.56 log genome equivalents (ge)/mL (N1) to 2.22 log ge/mL (IP4) when automated system was used; while values ranging between 2.08 (N1) and 2.34 (E) log ge/mL were observed when using column-based extraction method. Different targets were also evaluated in naturally contaminated wastewater samples with 91.2%, 85.3%, 70.6%, 79.4% and 73.5% positivity, for N1, N2, E, IP2 and IP4, respectively. Our benchmarked comparison study suggests that the aluminum precipitation method coupled with the automated nucleic extraction represents a method of acceptable sensitivity to provide readily results of interest for SARS-CoV-2 WBE surveillance.