Identification and genome sequencing of an influenza H3N2 variant in wastewater from elementary schools during a surge of influenza A cases in Las Vegas, Nevada

Real-time surveillance of infectious diseases at schools or in communities is often hampered by delays in reporting due to resource limitations and infrastructure issues. By incorporating quantitative PCR and genome sequencing, wastewater surveillance has been an effective complement to public health surveillance at the community and building-scale for pathogens such as poliovirus, SARS-CoV-2, and even the monkeypox virus. In this study, we asked whether wastewater surveillance programs at elementary schools could be leveraged to detect RNA from influenza viruses shed in wastewater. We monitored for influenza A and B viral RNA in wastewater from six elementary schools from January to May 2022. Quantitative PCR led to the identification of influenza A viral RNA at three schools, which coincided with the lifting of COVID-19 restrictions and a surge in influenza A infections in Las Vegas, Nevada, USA. We performed genome sequencing of wastewater RNA, leading to the identification of a 2021–2022 vaccine-resistant influenza A (H3N2) 3C.2a1b.2a.2 subclade. We next tested wastewater samples from a treatment plant that serviced the elementary schools, but we were unable to detect the presence of influenza A/B RNA. Together, our results demonstrate the utility of near-source wastewater surveillance for the detection of local influenza transmission in schools, which has the potential to be investigated further with paired school-level influenza incidence data.


H I G H L I G H T S G R A P H I C A L A B S T R A C T
• Wastewater surveillance can provide genomic information about influenza strains. • WGS can be performed to identify influenza H3N2 in wastewater. • Near-source wastewater surveillance for influenza should be paired with schoollevel incidence data. • Sequencing from wastewater can lead to identification of vaccine-resistant influenza sub-clades.

A B S T R A C T A R T I C L E I N F O Editor: Kyle Bibby
Keywords: SARS-CoV-2 COVID-19 Influenza Wastewater Elementary schools, mutation H3N2 Real-time surveillance of infectious diseases at schools or in communities is often hampered by delays in reporting due to resource limitations and infrastructure issues. By incorporating quantitative PCR and genome sequencing, wastewater surveillance has been an effective complement to public health surveillance at the community and building-scale for pathogens such as poliovirus, SARS-CoV-2, and even the monkeypox virus. In this study, we asked whether wastewater surveillance programs at elementary schools could be leveraged to detect RNA from influenza viruses shed in wastewater. We monitored for influenza A and B viral RNA in wastewater from six elementary schools from January to May 2022. Quantitative PCR led to the identification of influenza A viral RNA at three schools, which coincided with the lifting of COVID-19 restrictions and a surge in influenza A infections in Las Vegas, Nevada, USA. We performed Science of the Total Environment 872 (2023) 162058

Introduction
The global incidence of influenza cases and deaths was reduced to an unprecedented low level during the COVID-19 pandemic in 2020 and early 2021 (Dhanasekaran et al., 2022;Groves et al., 2022). Public health directives such as social distancing and masking, coupled with a reduction in travel, likely contributed to the reduced community transmission of influenza viruses (Fong et al., 2020); however, relaxation of COVID-19 restrictions during the 2021/2022 influenza season resulted in the circulation of a new influenza H3N2 subclade called 3C.2a1b.2a.2 (Melidou et al., 2022). Due to mutations in key antigenic sites on hemagglutinin (HA) of 3C.2a1b.2a.2, influenza vaccine effectiveness was found to be low during reported outbreaks in the 2021/2022 influenza season (Bolton et al., 2022;Melidou et al., 2022). The emergence and evolution of new influenza strains highlight the growing need to develop new surveillance programs that 1) detect and identify circulating strains to limit community transmission and 2) guide the development of efficacious vaccines.
Wastewater-based epidemiology (WBE) has been adapted widely during the COVID-19 pandemic to measure changing levels of SARS-CoV-2 (the virus that causes  and microbial communities at the community and facility-level (Ahmed et al., 2020;de Jonge et al., 2022;Farkas et al., 2022;Gerrity et al., 2021Gerrity et al., , 2022Harrington et al., 2022;Joshi et al., 2022;Kirby et al., 2022;Li et al., 2022;Nelson, 2022;Sellers et al., 2022;Vo et al., 2022bVo et al., , 2022cWolfe et al., 2022b). Infected individuals shed viral RNA through urine, stool, or saliva which enters sewer lines through toilets, sinks, and shower drains (Schmitz et al., 2021;Tiwari et al., 2021). When levels of SARS-CoV-2 RNA increase in sewage entering wastewater treatment plants (WWTPs), increasing case infections from communities served by WWTPs are observed. Similarly, when viral levels in wastewater decrease, a corresponding decrease in cases is observed, highlighting the utility of WBE as a real-time surveillance tool (Castro-Gutierrez et al., 2022;Godinez et al., 2022;Zdenkova et al., 2022). In addition to the quantification of viral levels in sewage, genome sequencing of nucleic acids extracted from wastewater has led to the identification of variants of interest (VOI) and concern (VOC) (Brumfield et al., 2022;Karthikeyan et al., 2022;Nemudryi et al., 2020;Reynolds et al., 2022;Tamáš et al., 2022;Vo et al., 2022a). As a result, wastewater sampling from treatment plants or buildings can provide operational information about the deployment of appropriate therapeutics and vaccines that target circulating variants.
Leveraging experience from the COVID-19 pandemic, we reasoned that community-scale wastewater monitoring programs could be used to detect influenza viruses in municipal wastewater. In addition, we queried whether our approach could be adapted to building-scale applications, specifically at elementary schools, since younger children are more susceptible to influenza and have high prevalence of symptomatic infections relative to other age groups (Ruf and Knuf, 2014;Sheu et al., 2016). Here, we aimed to interrogate 1) community and facility-level transmission of influenza in Southern Nevada through wastewater and public health surveillance and 2) genome sequencing of wastewater for the identification of viral variants. Our data highlight how wastewater programs can be deployed to detect local influenza transmission in schools and how genome sequencing from wastewater can lead to the identification of novel, vaccine-resistant influenza strains within this vulnerable subpopulation.

Wastewater sample collection and preparation
Wastewater surveillance at six elementary schools, each serving 200-800 students in Southern Nevada, was conducted between January 20th and June 13th of 2022. All six schools were located in the Las Vegas/Henderson school districts and were from similar socioeconomic and ethnicity (~39 % White,~32 % Hispanic,~11 % Black, and~10 % Asian) regions. Grab samples were collected from the elementary school manholes three days per week (on Mondays, Wednesdays, and Fridays) between 12:00 and 1:00 pm; three samples were collected every 5-10 min to generate a manual composite over 30 min of~500-1000mls. In addition, 10 L of grab primary effluent was collected at~10:00 am every Monday from a corresponding wastewater treatment plant (WWTP) serving approximately one million local residents and the schools in question (Gerrity et al., 2021).

Quantification of influenza A and B viral levels
For quantification and sequencing, nucleic acids were extracted from wastewater using the Wizard Enviro TNA Kit (Promega Cat #A2991) according to the manufacturer's instructions and eluted in 100 μL of RNasefree dH 2 O. First-strand cDNA was synthesized using the LunaScript RT SuperMix Kit (NEB). Quantification of influenza RNA in wastewater was performed using a Bio-Rad Opus qPCR instrument using CDC's Influenza SARS-CoV-2 multiplex assay (M gene for influenza A and NS2 gene for Influenza B) (CDC, 2020). Reactions, in triplicate, were performed in a final volume of 10 μL containing: 1 μL template cDNA, 5 μl of 2× KiCqStart Probe qPCR ReadyMix (Millipore-Sigma), 1 μL of 9 μM primer and 2.5 μM probe mix, and 3 μL of nuclease-free water. Primers and probes were purchased from Millipore-Sigma and Integrated DNA Technologies, respectively. Primer and probe details include: Influenza A forward 1 (5′-CAA GAC CAA TCY TGT CAC CTC TGA C-3′), Influenza A forward 2 (5′-CAA GAC CAA TYC TGT CAC CTY TGA C-3′), Influenza A reverse 1 (5′-GCA TTY TGG ACA AAV CGT CTA CG-3′), Influenza A reverse 2 (5′-GCA TTT TGG ATA AAG CGT CTA CG-3′), Influenza A probe (5′-FAM-TGCAGTCC T/ ZEN/CGCTCACTGGGCACG/IABkFQ-3′), Influenza B forward (5′-TCC TCA AYT CAC TCT TCG AGC G-3′), Influenza B reverse (5′-CGG TGC TCT TGA CCA AAT TGG-3′), Influenza B probe (5′-HEX-CCAATTCGA/ ZEN/ GCAGCTGAAACTGCGGTG/3IABkFQ-3′). Standard curves for influenza A were generated using a synthetic H3N2 RNA control (Twist Biosciences). Pepper mild mottle virus (PMMoV) was measured to determine the presence of human fecal content in the wastewater and used as a normalizing control. The concentration of PMMoV was determined by amplification of experimental samples along with dilutions of a PMMoV standard (provided as part of the GoTaq Enviro PMMoV Quant Kit from Promega) using quantitative reverse transcription PCR (RT-qPCR) with primers (forward: 5′-GAG TGG TTT GAC CTT AAC GTT TGA-3′; reverse: 5′-TTG TCG GTT GCA ATG CAA GT-3′) and probe (5′-TxRd-CCTACCGAAGCAAATG-BHQ2-3′) specific for PMMoV (Abdool-Ghany et al., 2022;Mondal et al., 2021). A no-template control was also included in each standard curve for the different targets. Cycling conditions included a denaturation step of 95°C for 2 min followed by amplification at 95°C for 5 s and 60°C for 30 s for 45 cycles. Public health surveillance for influenza in Las Vegas included data collected from local acute care hospitals and healthcare providers. Clinical cases were confirmed by PCR, culture, immunofluorescent antibody staining, IHC antigen staining, rapid influenza diagnostic tests with hospitalizations for 24 h or longer, or death certificates (SNHD, 2022). Given that influenza morbidity is not reportable to public health and that the majority of infected individuals suffer from mild forms of influenza, we expect that reported case counts represent an underestimation of total influenza infections.

Targeted genome sequencing and variant analyses
Whole genome sequencing libraries were constructed using the CleanPlex Respiratory Virus Research Panel from Paragon Genomics according to manufacturer's instructions. This panel combines SARS-CoV-2 whole genome amplification with 149 primers covering multiple gene segments for influenza A (H1N1,H1N2,H3N2), influenza B, and respiratory syncytial virus (RSV) types A and B. >10 ng of total RNA was processed for first-strand cDNA synthesis. Libraries were sequenced using an Illumina NextSeq 500 platform and a mid-output v2.5 (300 cycles) flow cell. Illumina adapter sequences were trimmed from reads using cutadapt v3.2. All sequencing reads were mapped to the respiratory pathogen genomes using bwa mem v0.7.17-r1188. Amplicon primers were trimmed from aligned reads using fgbio TrimPrimers v1.3.0 and segment 1 (hemagglutinin gene) consensus sequences were generated with iVar consensus v1.3. Genome coverages were calculated by samtools coverage (v1.10). The segment 1 consensus sequences were analyzed with Nextclade (v.1.8.1) and the phylogenetic tree was visualized on the Auspice webserver. Genome coverage of 6/8 influenza A genomic segments was >80 % with a median sequencing depth of >100-fold. However, the percentage of the full genome covered at 100-fold sequencing depth for all samples from the three schools was >55 % (Table S1). Raw fastq files are available on the National Center for Biotechnology Information website under BioProject PRJNA856656.

Human subjects statement
The Institutional Biosafety Committee (IBC) of the University of Nevada Las Vegas approved methods and techniques used in this study.

Results and discussion
Influenza incidence in Southern Nevada remained uncharacteristically low throughout the COVID-19 pandemic but started to increase in the late spring of 2022, peaking in week 15 (Fig. 1A). During this surge, the Southern Nevada Public Health Laboratory confirmed 403 hospitalized cases through public health surveillance, with 98.3 % positive for influenza A (Fig. 1B). At this time, the influenza vaccination rate was 21.3 % across all 2.4 million residents of the region and 25.1 % for the 0-10 age group (i.e., elementary school age) (DPBH, 2022). Notably, at least two contributing factors led to a late influenza season. During the COVID-19 surge between December 2021 and February 2022, cases of influenza were at historical lows due to social distancing and hygiene practices. Second, A.  Nevada lifted its mask mandate during week 6 of 2022, at which point face coverings were no longer required in indoor public places, including schools. Similar to other studies (Dumke et al., 2022;Mercier et al., 2022;Wolfe et al., 2022a), we asked whether wastewater could be analyzed to observe trends from influenza shedding-in this case at elementary schools-and confirmed cases at the community-level. To ensure that wastewater samples were processed appropriately, we also measured RNA levels of PMMoV-a plant virus that is found commonly in human fecal material. Across the sampling period, we detected PMMoV RNA at all six schools ( Fig. 1D and Suppl. Fig. 1), while influenza A viral RNA was detected at only three schools. Influenza RNA was not detected during the first three months of 2022, but influenza A was detected at School 1 at a concentration of 2.7 × 10 6 gene copies per liter (gc/L) in week 14, at School 2 at a concentration of 8.0 × 10 2 gc/L in week 21, and at School 3 at a concentration of 2.8 × 10 3 gc/L in week 22 (Fig. 1C). Interestingly, wastewater surveillance at the corresponding WWTP servicing the schools did not yield positive results for influenza A or for influenza B by qPCR, suggesting sensitivity limitations for influenza RNA detection at the community-scale (Heijnen and Medema, 2011). Corresponding clinical surveillance data indicated that detection of influenza A in elementary school wastewater occurred during increasing community transmission of influenza A in Las Vegas (Fig. 1C). This period of increased transmission occurred shortly after 1) the COVID-19 BA.1 surge ended, 2) lifting of a mask mandate, and 3) a time with relatively low influenza vaccination coverage.
We next performed tiled-amplicon sequencing and determined that H3N2 was present in the three influenza-positive wastewater samples (Fig. 2). Several unique mutations could be annotated across the genomes revealing that all belonged to the 2a.2 subgroup of the influenza A (H3N2) subclade 3C.2a1b.2a and demonstrating the feasibility of tracking influenza lineages through wastewater (Delahoy et al., 2021;Melidou et al., 2022;Wolfe et al., 2022a). Vaccines against influenza A (H3N2) are historically less efficacious than for other strains since H3N2 evolves more rapidly to escape immunity (Delahoy et al., 2021). In fact, the 2021-2022 influenza vaccines for the northern hemisphere were eventually updated to protect against both the 2a.1 and 2a.2 subgroups of the 3C.2a1b.2a subclade (Delahoy et al., 2021;Melidou et al., 2022). With this update, vaccinated children in Las Vegas may have been partially protected against 2a.2., but either due to low vaccination coverage or vaccine evasion, influenza A infections were still apparent based on clinical surveillance and detection of the viral RNA in wastewater at three of six elementary schools.
A limitation of this study was that it was not possible to assess health data directly from the elementary schools, including influenza vaccine coverage or confirmed cases at the time of the wastewater detections. Nonetheless, our study and others (Dumke et al., 2022;Mercier et al., 2022;Wolfe et al., 2022a) demonstrate the feasibility of leveraging wastewater to detect influenza outbreaks and identify subtypes at the buildinglevel. This could ultimately help public health officials mitigate the impacts of influenza outbreaks by more rapidly deploying interventions or assess the efficacy of future vaccines.
Supplementary data to this article can be found online at https://doi. Anthony Harrington, Van Vo, and Ching-Lan Chang are co-first authors and contributed equally.

Data availability
Data will be made available on request.

Declaration of competing interest
Dr. Carolina Koutras is an employee of R-Zero Systems. and the Southern Nevada Public Health Laboratory for their assistance with sample logistics and data access.