Evaluation of low-cost SARS-CoV-2 RNA purification methods for viral quantification by RT-qPCR and next-generation sequencing analysis: Implications for wider wastewater-based epidemiology adoption

Abstract

Based Epidemiology (WBE) consists of quantifying biomarkers in sewerage systems to derive real-time information on the health and/or lifestyle of the contributing population. WBE usefulness was vastly demonstrated in the context of the COVID-19 pandemic. Many methods for SARS-CoV-2 RNA determination in wastewater were devised, which vary in cost, infrastructure requirements and sensitivity. For most developing countries, implementing WBE for viral outbreaks, such as that of SARS-CoV-2, proved challenging due to budget, reagent availability and infrastructure constraints. In this study, we assessed low-cost methods for SARS-CoV-2 RNA quantification by RT-qPCR, and performed variant identification by NGS in wastewater samples. Results showed that the effect of adjusting pH to 4 and/or adding MgCl₂ (25 mM) was negligible when using the adsorption-elution method, as well as basal physicochemical parameters in the sample. In addition, results supported the standardized use of linear rather than plasmid DNA for a more accurate viral RT-qPCR estimation. The modified TRIzol-based purification method in this study yielded comparable RT-qPCR estimation to a column-based approach, but provided better NGS results, suggesting that column-based purification for viral analysis should be revised. Overall, this work provides evaluation of a robust, sensitive and cost-effective method for SARS-CoV-2 RNA analysis that could be implemented for other viruses, for a wider WEB adoption.

Keywords: COVID-19; RNA purification; SARS-CoV-2; Viral genomic surveillance; Virus quantification; Wastewater-based epidemiology.

Fig. 1

Effect of sample conditioning prior to filtration on viral RNA recovery. In all cases, total RNA was obtained using the NucleoSpin RNA Stool kit and estimated by RT-qPCR. A. Boxplots of SARS-CoV-2 RNA concentrations constructed with 30 data points per treatment, corresponding to 5 independent samples with 2 experimental replicates and 3 technical replicates (RT-qPCR estimations) each. B. Boxplots of Bovine Respiratory Syncytial Virus (BRSV) RNA concentration constructed with 20 data points per treatment, corresponding to 5 independent samples with 2 experimental replicates and 2 technical replicates (RT-qPCR estimations) each. Median (−), mean (♦), and outliers (○) are shown.

Fig. 2

Effect of physicochemical parameters in wastewater samples on SARS-CoV-2 RNA recovery. A. Cullen and Frey Diagram showing the type of distribution followed by SARS-CoV-2 RNA concentration data. B. Spearman Correlation between pH, temperature (T), ionic conductivity (C), chemical oxygen demand (COD) in wastewater, and SARS-CoV-2 RNA concentration. In both cases n = 123. Asterisks (*) indicate that correlation coefficient is statistically significant.

Fig. 3

Effect of RNA purification method from wastewater samples on SARS-CoV-2 RNA N1 quantification by RT-qPCR. Boxplots were constructed with 63 data points: 21 independent samples per treatment with 3 RT-qPCR estimations each. Median (−), mean (♦), and outliers (○) are shown.

Fig. 4

Effect of standard curve material on the quantification of SARS-CoV-2 RNA in wastewater samples. A total of 9 independent wastewater samples were assayed in triplicate by RT-qPCR. SARS-CoV-2 RNA N1 concentration ratios for each sample are plotted against SARS-CoV-2 RNA N1 concentration estimated with the plasmid dsDNA standard curve. Ratios were calculated by dividing SARS-CoV-2 RNA N1 concentration obtained using plasmid dsDNA over linear dsDNA as standard. Insert, Boxplot distribution of N1 RNA concentration ratios.

Fig. 5

Effect of RNA extraction method on SARS-CoV-2 NGS reads' numbers in wastewater samples. A total of 24 RNA samples for each extraction method were sequenced and the total number of reads at each step and that mapped to the GISAID reference database are shown.

Fig. 6

SARS-CoV-2 lineage detection from wastewater RNA samples isolated with the NucleoSpin RNA Stool kit and a modified TRIzol method. Total RNA from 24 wastewater samples were isolated with either method and sequenced through NGS. Heatmap was constructed using GISAID-matched read count, in logarithmic scale. Numbers and dates below correspond to each sample analyzed with both methods, and sampling dates. Low-detection lineages (<5% abundance of GISAID-matched reads) were not considered in the heatmap. Alpha (B.1.1.7 and Q lineages), Beta (B.1.351 and descendent lineages), Gamma (P.1 and descendent lineages), Delta (B.1.617.2 and AY lineages), Lambda (C.37, C.37.1) and Mu (B.1.621, B.1.621.1) variants are represented in this graph.