2.5. Quantification of SARS-CoV-2 N1 and N2 genes, viral surrogates and PMMoV by RT-qPCR

All RNA extracts of the longitudinal samples were analyzed by RT-qPCR for four viral targets: The N1 and N2 gene targets of SARS-CoV-2, the surrogate virus and PMMoV. All N1, N2 and MHV analyses, as well as PMMoV analyses for Lugano and Lausanne were performed at EPFL. PMMoV and Sendai virus analyses for Zurich were performed at Eawag. The samples to test surrogate virus recovery were analyzed at EPFL for three viral targets: MHV, Sendai virus and Φ6. To detect the presence of SARS-CoV-2 RNA, the CDC N1 and N2 assays were used (^{Lu et al., 2020}). PMMoV and MHV were analyzed by previously reported assays (^{Besselsen et al., 2002; Haramoto et al., 2013; Zhang et al., 2006}). The design for primers and probes for Φ6 were adapted from ^{Gendron et al. (2010)} according to the suggestion of Heather Bischel (University of California, Davis). For Sendai virus, primers and probes were designed for the purpose of this project. A summary of all primers and probes and the RT-qPCR protocols is given in the Supporting Information (Table S1).

To calibrate the different RT-qPCR assays, standard curves for each viral target were generated using either double-stranded DNA gblocks gene fragments (viral surrogates and PMMoV), or a 2019-nCoV_N positive control plasmid (cat nº 10006625, SARS-CoV-2 N1 and N2). Both gblocks and plasmids were purchased from Integrated DNA Technologies (Coralville, IA, USA).

RT-qPCR amplifications were performed in 25 µl reactions using RNA UltraSense™ One-Step Quantitative RT-PCR System (cat nº 11732–927, Invitrogen, Carlsbad, CA, USA), amended with 4 µl of bovine serum albumin (2 mg/ml; cat. nº 10711454001, Sigma-Aldrich, Buchs, Switzerland) on a Mic qPCR Cycler (Bio Molecular Systems, Upper Coomera, Queensland, Australia). In each RT-qPCR reaction, 5 µl of RNA extract or calibration standard were used. For PMMoV, the RNA extract was diluted 1:10 prior to RT-qPCR analysis. All RT-qPCR runs included no-template controls and negative extraction controls to monitor for contamination during the extraction and amplification process. The preparation of PCR mastermix and standards, as well as sample loading were performed in separate locations to avoid contamination. Cq determination was performed using the micPCR software (v2; Bio Molecular Systems).

RT-qPCR limits of detection (LOD) were determined as the lowest concentration (N1, N2 and MHV) or the lowest standard (PMMoV, Sendai, Φ6) with a 95% or greater detection probability. The limit of detection (LOD) for each gene target were determined from pooled standard curves (n ≥ 3) in R using the Generic qPCR Limit of Detection (LOD) / Limit of Quantification (LOQ) calculator (^{Merkes et al., 2019}). Samples with a measurable RT-qPCR signal < LOD were assigned the concentration of the LOD of the respective assay. Samples which yielded no detectable RT-qPCR signal were set to the theoretical minimal LOD (3 gc/reaction) (^{Ståhlberg and Kubista, 2014}).