reverse osmosis system

Science of the Total Environment 646 (2019) 427–437

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Noroviruses in raw sewage, secondary effluents and reclaimed water produced by sand-anthracite filters and membrane bioreactor/reverse osmosis system Tatiana Prado a,⁎, Antônio de Castro Bruni a, Mikaela Renata Funada Barbosa a, Suzi Cristina Garcia a, Luisa Zanolli Moreno b, Maria Inês Zanoli Sato a a Department of Environmental Analysis, Division of Microbiology and Parasitology, Environmental Company of the São Paulo State (CETESB), Av. Prof. Frederico Hermann Jr., 345, São Paulo, SP 05459-900, Brazil b Laboratory of Molecular Epidemiology and Antimicrobial Resistance, School of Veterinary Medicine and Animal Science, University of São Paulo, Av. Prof. Dr. Orlando Marques de Paiva, 87, São Paulo, SP 05508-270, Brazil

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

• Noroviruses are widely disseminated in the municipal wastewaters of São Paulo city. • Higher viral loads were observed in winter season. • Activated sludge processes showed higher impact on NoV GII removal than NoV GI. • Results demonstrated the best performance of the MBR/RO systems in virus removal. • Prevalence of GII.17 indicates a shift in the epidemiology of NoV GII in the city.

a r t i c l e

i n f o

Article history: Received 19 May 2018 Received in revised form 20 July 2018 Accepted 21 July 2018 Available online 23 July 2018 Editor: Frederic Coulon Keywords: Membrane bioreactor-MBR Noroviruses Reclaimed water Reverse osmosis Sand-anthracite filters Wastewater treatment plants

a b s t r a c t The importance of noroviruses (NoVs) in the epidemiology of waterborne diseases has increased globally in the last decades. The present study aimed to monitor genogroup I and II noroviruses in different treatment stages of four wastewater treatment plants (WWTPs) in the metropolitan São Paulo. WWTPs consist of secondary (activated sludge) and tertiary treatments (coagulation, sand-anthracite filters, membrane bioreactor (MBR)/reverse osmosis (RO) and chlorination). Raw sewage (500 mL) and treated effluents (1 L) were concentrated by celite and reclaimed water (40 L) by hollow-fiber ultrafiltration system. Quantitative (qPCR) and nested PCR with nucleotide sequencing were used for quantification and molecular characterization. NoVs were widely distributed in raw wastewater samples (83.3%–100% NoV GI and 91.6%–100% NoV GII) and viral loads varied from 3.8 to 6.66 log10 gc L−1 for NoV GI and 3.8 to 7.3 log10 gc L−1 for NoV GII. Mean virus removal efficiencies obtained for activated sludge processes ranged from 0.3 to 0.8 log10 for NoV GI and 0.4 to 1.4 log10 for NoV GII. NoVs were not detected in the reuse water produced by MBR/RO system, while sand-anthracite filters resulted in a NoV GI and GII decay of 1.1–1.6 log10 and 0.7–1.6 log10, respectively. A variety of genotypes (GI.2, GI.3a, GI.3b, GI.5, GII.1, GII.4 Sydney 2012, GII.5, GII.6, GII.17) was observed, with a predominance of GI.2 and GII.17 in the different genogroups. These results corroborate with recent data about the entry and dissemination of the emerging genotype GII.P17-GII.17 Kawasaki 2014 in the country, and may indicate a change in the epidemiological patterns

⁎ Corresponding author. E-mail address: [email protected] (T. Prado).

https://doi.org/10.1016/j.scitotenv.2018.07.301 0048-9697/© 2018 Elsevier B.V. All rights reserved.

428

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

of norovirus strains circulation in this region. This is the first large-scale study to evaluate burden and genotypes of noroviruses in WWTPs in Brazil, providing a rapid diagnosis of viruses circulating in the population. © 2018 Elsevier B.V. All rights reserved.

1. Introduction Noroviruses (NoV) are currently considered the main etiological agents in the world causing outbreaks of acute gastroenteritis associated with water and food (Kirk et al., 2015). Annually, they are estimated to cause 684 million episodes of diarrheal disease and 212,000 deaths among all ages and from all modes of transmission. The disease results in an economic burden of US $ 4.2 billion in direct health system costs and US $ 60.3 billion in societal costs (Bartsch et al., 2016). In Latin America, NoV are associated with almost 1 out of every 6 hospitalizations due to acute diarrhea in children younger than 5 years of age (O'Ryan et al., 2017). In metropolitan São Paulo, NoV is the main cause of foodborne diseases outbreaks (25%) in total outbreaks investigated between 2012 and 2016 (58/231) (Coordination of Health Surveillance State of SP SINAN NET/COVISA/CCD/Banco EPI/Surtos DTA) (personal communication). NoVs are single-stranded RNA viruses, positive polarity, not enveloped, and classified within the genus Norovirus, from the family Caliciviridae (De Graaf et al., 2016; Vinjé, 2015). The viral particle is 27 to 40 nm in diameter and 7.5 to 7.7 kb in genome length, containing 3 open reading frames (ORFs) (ORF1, ORF2 and ORF3) (Vinjé, 2015). NoV is currently divided into 6 genogroups, designated GI to GVI, based on the amino acid identity of the largest structural protein (VP1) (De Graaf et al., 2016). Genogroups I and II are responsible for the majority of human disease cases, while GIV is rarely detected as a cause of sporadic or epidemic disease (Vinjé, 2015). To date, 9 genotypes have been recognized in GI and 19 in GII (Vinjé, 2015). Among the established genotypes, NoV GII.4 strains are prevalent in outbreaks of gastroenteritis around the world, including Brazil (Fioretti et al., 2011; Vinjé, 2015). Due to evolution, new pandemic variants of NoV GII.4 have emerged every 2 to 3 years since the mid-1990s, replacing previously dominant strains of GII.4, often, but not always, associated with an increased number of outbreaks (Vinjé, 2015). Norovirus is stable in the environment, and has an infectious dose of 50% (ID50) reaching 18 to 103 genome copies (Lopman et al., 2012; Vinjé, 2015). The ease with which noroviruses are transmitted and the low infectious dose required to establish an infection result in extensive outbreaks. The risk of NoV infection associated with environmental exposure, particularly to water contaminated with human fecal matter and reclaimed water, is an emerging research topic (Eftim et al., 2017; Sano et al., 2016). Water reuse has been considered an environmentally sound and sustainable alternative to mitigate the effects of hydric stress already verified in several parts of the world (Sano et al., 2016). However, a matter of considerable societal concern is the potential risk to human health associated with human contact with waterborne pathogenic microorganisms present in wastewater (Sano et al., 2016; Yasui et al., 2017). More specifically, waterborne viral pathogens are inadequately removed from existing wastewater treatment systems and bacterial indicators used to assess water quality fail to detect their presence accurately (Montazeri et al., 2015; Sano et al., 2016). Several researches have focused on the choice of suitable viral indicators to evaluate human fecal contamination in different aquatic ecosystems, including human adenoviruses (HAdV), JC polyomaviruses, pepper mild mottle viruses (PMMoV) and bacteriophages (Amarasiri et al., 2017; Fumian et al., 2013; Kitajima et al., 2014; Prado et al., 2018; Sidhu et al., 2017). Noroviruses, although are not considered good viral markers to evaluate human fecal

contamination in aquatic ecosystems, are important pathogens associated to food and waterborne diseases. Therefore, in order to reduce risks to human health associated with discharge of final effluents into the environment and to address the growing pressure for direct or indirect reuse of reclaimed waters, methods for the detection, removal and prediction of concentrations of viral pathogens need to be established and optimised. A recent work using meta-analysis to characterize the NoV load in raw sewage from WWTPs of different parts of the world included survey data from several continents: Asia, Europe, USA and New Zealand, but none specifically from South America (Eftim et al., 2017). In Brazil, there are few relevant studies evaluating the load and efficiency of sewage treatment systems for NoV removal (Fumian et al., 2013; Victoria et al., 2010) and none specifically for tertiary or advanced treatment systems, including bioreactor membrane (MBR). Thus, there is a lack of information on the performance of conventional and advanced sewage treatment systems operating in real scale in the country and the virological quality of final effluents produced. Reclaimed water generated in Municipal Wastewater Treatment Plants (WWTPs) has been used in São Paulo city mainly for industrial and urban purposes, including landscape irrigation, street and vehicle washing, civil construction, firefighting, among others. The volume produced is sold to Municipalities, construction companies and industries in the textile, petrochemical and paper and pulp sectors. There is also an expectation to expand the production of reuse water in the coming years, including reuse in the recharge of aquifers. Despite the absence of more specific national legislations to regulate the practice in this sector, recently the State of São Paulo implemented legislation (Resolution SES/SMA/SSRH No. 01 - June 28, 2017) establishing standards for non-potable direct reuse of water from WWTPs for urban purposes. In this legislation, monitoring and water quality standards were established for reuse with moderate and severe restriction. Limits of thermotolerant coliforms (≤200 CFU/100 mL) and helminth eggs (1 egg/L) were included for severe restriction. Moderate restriction requires a treatment process that guarantees no measurable levels of pathogens, including viruses. Although, the legislation does not specify the viruses to be investigated, the competent authorities to enforce this regulation will require analyses of pathogenic viruses or viral indicator, such as bacteriophages or adenoviruses, to characterize the efficiency of reuse water treatment process. Therefore, the aim of this study was to monitor NoV GI and GII in urban wastewaters, treated effluents and reclaimed water samples produced at four municipal WWTPs in the metropolitan region of São Paulo, during a one-year period (April 2015 to March 2016). WWTPs employ secondary (conventional activated sludge) and diverse tertiary treatments (coagulation/sedimentation/microfiltration/MBR/reverse osmosis system) and final disinfection by chlorination. Besides the viral loads detected in distinct stages of treatment processes, genotypes were identified through phylogenetic analysis, providing important information on virus circulation in the population served by sewage systems. 2. Material and methods 2.1. Sampling and characteristics of WWTPs Grab samples were collected in different sewage treatment process stages from four WWTPs that receive sewage of approximately 8 million people living in São Paulo, the largest city in South America. The WWTP

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

429

characteristics are shown in Table 1 (modified from Prado et al., 2018). These Plants consist of primary treatment (sedimentation), secondary treatment (activated sludge) and tertiary treatment (filtration or ultrafiltration/reverse osmosis, and chlorination) (Table 1). Samples of 500 mL of raw sewage, 1 L of secondary effluent and 40 L of reclaimed water were collected into sterile polypropylene bottles, once a month between April 2015 and March 2016, totalizing 144 samples. Reclaimed water samples were dechlorinated with sodium thiosulphate (10%) after collection. The samples were maintained under refrigeration (4 °C) for transportation and processed within 24 h.

7550-30) was used and silicone hoses were coupled to the system to perform the filtration at a flow rate of approximately 800 mL/min. The permeated polyphosphate solution was discarded. After sample filtration, viruses were eluted by reverse flow with 500 mL of 0.1% Tween 80, 0.01% NaPP and 0.001% Antifoam Y-30 (backwash solution) and submitted to a secondary concentration by adsorption to celite, as described above. High virus recovery rates have been achieved by hollow-fiber ultrafiltration followed by reconcentration using celite and seems to be promising to recover enteric viruses from reuse waters (Rhodes et al., 2011).

2.2. Physicochemical parameters

2.4. Viral genomic extraction and reverse transcription (RT) reaction

Physical and chemical parameters were measured to obtain complementary information about the wastewater characteristics through the treatment stages and the possible influence of these parameters in the viral dynamic in the WWTPs (for details, consult Table S1 - Supplementary material). Water temperature, pH, conductivity and residual chlorine were measured in the field and, total organic carbon, total solids and turbidity were analyzed at CETESB Chemistry Laboratory. Analyses were performed according to the Standard Methods (APHA, 2012) recommended methodologies.

Viral RNA was extracted from concentrated samples (140 μL of eluate) using the QIAmp Viral RNA Mini Kit (Qiagen, Inc., Valencia, CA) according to the manufacturer's instructions. The cDNA synthesis was carried out using 10 μL of extracted RNA, random primers (pd(N)6; 50A260 units; Amersham Biosciences, Chalfont St Giles, Buckinghamshire, UK) and Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA).

2.3. Concentration methods

Quantitative polymerase chain reactions (qPCRs) were performed using StepOne Plus® Real-Time PCR System (Applied Biosystems) with TaqMan detection. The primers set, probes and reaction conditions were previously described (USEPA, 2014) (Table 2). Amplification was performed in a total volume of 20 μL/reaction containing 14 μL of kit TaqMan® Environmental Master Mix 2.0 (Applied Biosystems, Foster City, CA, USA), 6 μL of cDNA and the respective concentrations of primers and probes as described in Table 2. The reaction was run under the following conditions: incubation at 50 °C for 2 min for activation of uracil-N-glycosylase, initial denaturation at 95 °C for 10 min, and 45 cycles of 15 s at 95 °C, followed by 1 min at 60 °C. Quality controls included the use of different rooms (reagent preparation, sample preparation and, amplicon and product detection) to prevent contamination between samples and from previous PCR amplicons generated in the laboratory, barrier-filtered pipette in all manipulations, and positive and negative (non-template controls - NTC)

Virus concentration in raw and treated sewage was carried out according to a protocol previously described by Brinkman et al. (2013), since these authors have reported high recovery rates (104%) for noroviruses in sewage. Volumes of 500 mL of sewage sample and 1 L of secondary sewage were used to concentrate viruses using celite (diatomaceous earth) (Brinkman et al., 2013). After procedures, the concentrated samples (̴ 30 mL) were kept at −80 °C until further analysis. Reclaimed water samples (40 L) were concentrated by ultrafiltration (Hollow-Fiber Ultrafiltration - UF) using a dead-end UF system (DEUF) as described by Smith and Hill (2009). The polysulfone filter (Dialyzer Fresenius Optiflux F200NR - Hemoflow), with 30KDa pore and surface area of 2 m2, was pretreated immediately prior to the experiments with the addition of 1 L of a polyphosphate solution (0.01% NaPP in deionized water). A peristaltic pump (Cole-Parmer Masterflex L/S - Model

2.5. Quantitative qPCR protocols for norovirus detection

Table 1 Primary, secondary and tertiary treatment processes applied at the four wastewater treatment plants selected in this study. Wastewater treatment process

WWTP-1

WWTP-2

WWTP-3

WWTP-4

Inflow (m3 s−1)a Population equivalent Preliminary treatment

9.7 4.4 million Screens and grit removal Primary settlement (±1 h) Activated sludge (AS) (±8 h) + secondary settlement Sand-anthracite filter (20 μm)

2.5 1.2 million Screens and grit removal

0.8 720 thousand Screens and grit removal Primary settlement Activated sludge (AS) + secondary settlement

1.9 1.4 million Screens and grit removal

Primary treatment (HRT) Secondary (biological) treatment (HRT)

Tertiary treatment (reclaimed water)

Disinfection

Reclaimed water uses

Primary settlement Activated sludge (AS) (±6–10 h) + secondary settlement (2–5 h)

Coagulation (Aluminum polychloride Aln-(OH)m(Cl3)n-m), flocculation/sedimentation, Sand-anthracite/zeolites filter (20 μm) Sodium Sodium hypochlorite (Ct = 60–120 min, hypochlorite (Ct = 15–20 ppm (TRC)) ≥30 min, 15–20 ppm (TRC)) Restricted urban Restricted urban use and industrial use use

Ct = contact time; HRT = hydraulic retention time; TRC = total residual chlorine. a Mean inflow rate provided by the Sanitation Company.

Primary settlement (±2 h) Activated sludge (AS) (±4–6 h) + secondary settlement (±3–5 h)

Sand-anthracite filter (20 μm)

Disk filter (ARKAL) (400 μm), anoxic chambre (N and P removal), membrane bioreactor - MBR (pore size of 0.05 μm) - submerged membranes, reverse osmosis (b0.001 μm)

Sodium hypochlorite (Ct = 30 min, 5–15 ppm (TRC)) Restricted urban use

Chlorine dioxide (residual chlorine: 0.2 ppm).

Industrial use (boilers, cooling water), urban use (fire fighting system, street washing)

430

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

Table 2 Primers and probes used in qPCR assays, target on the genome, sequence and final concentrations. NoV Primers and probes

Target region

Sequence (5′ - 3′)

GI

NorGIBF NorGIBR NorGIBP

ORF2

GII

NorGIIF NorGIIR NorGIIP

ORF2

CGCTGGATGCGNTTCCAT CCTTAGACGCCATCATCATTTAC 6 FAM-TGGACAGGA GAYCGCRATCT-TAMRA ATGTTCAGRTGGATGAGRTTCTCWGA TCGACGCCATCTTCATTCACA 6 FAM-AGCACGTGGGAGGGCG ATCG-TAMRA

Final concentrations 500 nM 900 nM 250 nM 500 nM 900 nM 250 nM

Degenerate bases in primers and probes are as follows: N equals a mixture of all four nucleotides; R equals A + G; Y equals T + C; W equals A + T. FAM, 6-carboxyfluorescein; TAMRA, 6-carboxytetramethylrhodamine.

controls for each set of amplifications. The positive controls, NoV GI (cod C3121) and GII.4 (cod C16), were obtained from fecal suspensions of clinical samples kindly provided by Dr. Rita de Cássia Compagnoli Carmona from Adolfo Lutz Institute (IAL), São Paulo State Health's Secretary. Synthetic DNA oligonucleotides (gBlock® Gene Fragment), commercialized by Integrated DNA Technologies (IDT, Lowa, USA), were used to establish the standard curve. The amplicon size corresponded to 300 bp, targeting the region of the ORF2 of the genome flanked by the primers: nucleotide (nt) position 5161 to 5460 - reference strain Norwalk virus M87661 (NoV GI) and nt position 4861 to 5160 - reference strain Hu/Houston EU310927 (NoV GII). Serial dilution (10× fold) of gBlock®, containing the target sequences of each virus type, was used to build the standard curves, reaching 108–101 Log10 copies per reaction in each amplification cycle (CycleThreshold - Ct) (Fig. S1 A,B - supplementary material). A standard curve was generated by linear regression of the relationship between the quantification cycle value and copy number. Amplification data were collected and analyzed using StepOne Software version 2.2.2 (Applied Biosystems, CA, USA). Samples with signals that crossed the threshold line, presenting a characteristic sigmoid curve and values of Ct reaching the maximum value of 40 (Ct ≤ 40), were considered positive. We assumed that the quantification limit was 3.0 genome copies (gc) per μL as the lowest standard dilution within the linear dynamic range of the standard curve. However, samples with no increase in fluorescence after 40 cycles were considered ‘below the detection limit’. All amplification reactions (samples, positive and negative controls) were carried out in duplicate and, in triplicate for the experiments to evaluate the efficiency of the standard curves (Fig. S2 A,B - Supplementary material). All standard curves presented a linear correlation coefficient above 0.999. The descriptive statistic, involving calculations of linear correlations and reaction efficiency are shown in Fig. S2 (Supplementary material). Results were expressed as genome copies per liter (gc L−1), considering the volumes of the sample, the concentrate (eluate), the RNA extracts and the qPCR reaction.

2.6. Internal process control (PP7 bacteriophage) The performance of the virus recovery methodology was monitored by the incorporation of PP7 bacteriophage (ATCC 15692-B2) as an internal process control (IPC). PP7 viral stocks were prepared according the protocol described by Rajal et al. (2007), employing the host Pseudomonas aeruginosa (ATCC 15692). All samples were spiked with 1 mL of PP7 bacteriophage stocks previously quantified with qPCR assay (Fumian et al., 2010) using primers and probe previously described (Rajal et al., 2007). Virus recovery rates were calculated according to the following

eq. (E):
 h i Eð%Þ ¼ no :genome copies gc L−1 of PP7 after concentration=no :gc of spiked PP7 100

Recovery rates were not employed to estimate norovirus concentrations in the samples by adjusting the observed viral genome copy numbers. To evaluate the presence of inhibitors, the nucleic acids extracts from all samples (n = 141) were also tested with a 10-fold dilution for bacteriophage PP7. Samples that presented inhibition or resulted negative for norovirus detection were retested, diluting nucleic acids extracts (10-fold dilution). 2.7. Nested PCR and noroviruses sequencing Samples that were positive for noroviruses by qPCR were further amplified through PCR prior to sequencing. Nested PCR was performed as described by Kitajima et al. (2010), with forward and reverse primers towards ORF1-ORF2 junction and 5′ region of VP1 gene (region C), respectively. Nested PCR products were separated on 2.0% electrophoresis- grade agarose gel (GIBCO BRL, Life Technologies, Inc., Grand Island, NY) and stained with GelRed™. Images were obtained using an image capture system (UVP Life Science Software, CA, USA). Subsequently, they were sent to nucleotide sequencing. Sequencing reactions were performed commercially and the amplicons were purified with ExoSAP-IT (GE Healthcare UK Ltd., Buckinghamshire, UK) and sequenced using the Big Dye® kit Terminator Cycle Sequencing Kit v.3.1 (Applied Biosystems, CA, USA) on an ABI Prism 3130xl Genetic Analyzer® (Applied Biosystems, CA, USA), as described by Otto et al. (2008). Nucleotide sequences were edited and aligned with Bioedit® Sequence Alignment Editor (v.7.0) (Hall, 1999). The sequences were compared with their respective prototypes as well as with other sequences available in the National Center for Biotechnology Information (NCBI/ GenBank) (http://www.ncbi.nlm.nih.gov/) and genotypes were assigned using Norovirus Automated Genotyping Tool (Kroneman et al., 2011). 2.8. Phylogenetic analysis Phylogenetic analysis was carried out with MEGA software v.5.2.2 (Phoenix, AZ) (Tamura et al., 2011) using the Maximum Likelihood method with Kimura two-parameter plus Gamma distribution model. A total of 2000 pseudoreplicates were used for branch support statistical inference. Sequences obtained in this study were submitted to the GenBank Database under the accession numbers: MF784760 to MF784801 (NoV GI), and MF683014 to MF683039 (NoV GII). 2.9. Statistical analysis The logarithmic transformation was applied to norovirus GI and GII concentrations. In addition, positivity indicator variables were created for norovirus GI and GII. For statistical analysis, concentrations below the limits of detection (LOD), left-censored observations, were assigned a value equal to the LOD divided by square root of 2 (Croghan and Egeghy, 2003). The LODs were calculated for each analyzed sample and they were not fixed, considering that different volumes were obtained for each sample after the concentration step. Box Plots with Notches option were produced in descriptive analysis of data and figures were produced to illustrate the differences between medians (Chambers et al., 1983). Generalized Linear Models (Nelder and Wedderburn, 1972) were adjusted using the variables indicating positivity of Norovirus GI and

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

GII. WWTP and the treatment stage were used as fixed factors and the physicochemical parameters were the covariates of the model. The binomial distribution was applied. Generalized linear mixed models (Bates and Debroy, 2004) were adjusted to the concentration data of NoV GI and GII. WWTP and treatment stage were the fixed factors and the physicochemical parameters were the covariates of the model. When adjusting statistical models, it is usual to consider two significance levels: one for the model itself and one for the model parameters. In this study were considered p ≤ 0.001 and 0.20, respectively (Bursac et al., 2008). 3. Results 3.1. Noroviruses in raw sewage, treated effluents and reclaimed water samples Over the study period, 141 wastewater samples (48 raw sewage, 48 secondary effluents and 45 reclaimed water samples) from four WWTPs at São Paulo Metropolitan Region were collected and analyzed. Bacteriophage PP7 used as an IPC was detected in 85.4% (41/48) of all raw sewage samples evaluated in this study, while for secondary effluents and reclaimed water samples the percentage of detection was 95.8% (46/ 48) and 100% (45/45), respectively. Table 3 shown mean recovery rates for PP7 obtained during all monitoring program. NoV GI was detected in 47.9% (69/144) of the samples, while the detection rate for NoV GII was 59.0% (85/144) using the qPCR assay (Table 4). NoV GI and GII are widely distributed in raw sewage from São Paulo city, with detection rates varying from 83.3% (WWTP-4) to 100% (WWTPs 2 and 3) for NoV GI and from 91.6% (WWTPs 4) to 100% (WWTPs 1, 2 and 3) for NoV GII (Table 4). Lower detection rates were observed in secondary and tertiary treatments (Table 4). Statistical analyzes demonstrated that detection rates of NoV GI were equivalent in all WWTPs, while for NoV GII, significantly higher detection rates were found for WWTP-1 at all treatment stages (p = 0.024), being observed an association with greater number of equivalent population in this plant. Peaks of NoV concentrations in raw sewage could be observed during the winter months (July/August) for genotype GI and GII, with seasonal concentrations significantly higher in winter (Fig. 1). The Fig. 2 shows median concentrations of Norovirus GI and GII obtained at different stages of WWTPs. In general, median viral loads of NoV GII in raw sewage were higher than NoV GI (Fig. 2A and B). Median concentrations of NoV GI in raw sewage reached 5.2 log10 (WWTP-1), 5.3 log10 (WWTP-2), and 5.4 log10 in WWTP-3 and 4 (Fig. 2A). In the secondary effluents, the median loads of NoV GI were: 4.8 log10 (WWTP-1), 4.9 log10 (WWTP-2), 5.1 log10 (WWTP-3) and 4.6 log10 (WWTP-4) (Fig. 2A). For reclaimed water samples, median viral loads of 3.5 log10 were obtained for NoV GI in WWTPs 1 and 2 (Fig. 2A). For NoV GII, medians of 5.4 log10 (WWTP-1), 5.6 log10 (WWTP-2), 5.5 log10 (WWTP-3) and 5.4 log10 (WWTP-4) were obtained in raw sewage (Fig. 2B). For secondary effluents the medians ranged 4.4 log10

431

Table 4 Prevalence and viral loads (log10 gc L−1) of NoV GI and NoV GII in 141 samples from four WWTPs. No. of samples analyzed

Viral prevalence (%)

Viral load (log10 gc L−1) (range: min–max)

NoV GI

NoV GII

NoV GI

NoV GII

12 12 12 12

91.6 100 100 83.3

100 100 100 91.6

4.34–6.25 3.94–6.66 4.14–6.09 3.8–6.26

4.92–7.31 4.61–6.57 4.32–6.23 3.84–6.28

Secondary effluents WWTP-1 12 WWTP-2 12 WWTP-3 12 WWTP-4 12

58.3 58.0 0 58.3

91.6 66.6 25.0 66.6

4.07–5.04 3.47–5.1 ND 3.64–4.91

3.11–5.35 3.23–4.54 3.53–5.29 3.55–5.5

Reclaimed water WWTP-1 09 WWTP-2 12 WWTP-3 12 WWTP-4 12

22.2 8.3 0 0

55.5 33.3 0 0

2.29–3.55 2.66 ND ND

1.99–2.99 2.18–2.49 ND ND

Stages of the WWTPs

Raw sewage WWTP-1 WWTP-2 WWTP-3 WWTP-4

ND = not detected (below of limit of detection - LOD). The LODs calculated for raw sewage, secondary effluent and reclaimed water were as following (mean ± SD) (gc L−1): (3.5 × 105 ± 5.5 × 104), (1 × 105 ± 5.1 × 104), (3.4 × 103 ± 1.0 × 103) (NoV GI); (4.0 × 105 ± 0.0), (1.1 × 105 ± 3.7 × 104), (3.1 × 103 ± 1.1 × 103) (NoV GII).

(WWTP-1), 4.2 log10 (WWTP-2), 5.1 log10 (WWTP-3) and 4.7 log10 (WWTP-4) (Fig. 2B). For reclaimed water samples, NoV GII were only detected in WWTPs 1 and 2, with median viral load of 3.0 and 3.5 log10, respectively (Fig. 2B). The median log10 reduction values achieved by activated sludge processes and tertiary treatments and disinfection are shown in Table 5. Activated sludge processes were capable to reduce an average of 0.3 to 0.8 log10 of NoV GI and 0.4 to 1.4 log10 of NoV GII (Table 5). An additional reduction in concentrations of NoV GI and GII could be achieved by tertiary treatments (Table 5). Total LRV ranged from 1.7 to 1.9 log10 for NoV GI and from 1.9 to 2.4 log10 for NoV GII (Table 5).

Table 3 Bacteriophage PP7 recovery (%) at the different treatment process stages of the four WWTPs. WWTPs

1 2 3 4 a b c

PP7 recovery (Mean ± SDa) Raw sewageb

Secondary effluentb

Reclaimed waterc

30.9 ± 32.3 24.3 ± 15.7 29.7 ± 26.03 28.7 ± 24.02

46.5 ± 51.1 34.7 ± 26.2 62.9 ± 54.7 62.1 ± 54.1

26.8 ± 27.5 39.4 ± 36.7 59.9 ± 52.09 32.7 ± 47.2

Standard deviation. Celite concentration. HFUF + celite concentration.

Fig. 1. Boxplot representing seasonal variation of noroviruses density (median concentrations − log10 gc L−1) in raw sewage samples collected during April 2015 to March 2016. The whiskers extend to the most extreme data point no N1.5 times the interquartile (25th–75th) range from the box.

432

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

Fig. 2. Boxplots Notched showing median concentrations (log10 gc L−1) of noroviruses GI (A) and GII (B) in different stages of the WWTPs with the 25th–75th percentile values. The whiskers extend to the most extreme data point no N1.5 times the interquartile (25th–75th) range from the box. Left-censored data set divided by square root of 2 were included in analysis. RS = raw sewage; SE = secondary effluent; RW = reclaimed water.

3.2. Phylogenetic analysis Table 5 Median log10 reduction values (LRV) obtained by different types of treatment. Median LRV AS

Tertiary treatment

Total LRV

WWTP-1 NoV GI NoV GII

0.4 1.0

1.3 1.4

1.7 2.4

WWTP-2 NoV GI NoV GII

0.4 1.4

1.4 0.7

1.8 2.1

WWTP-3 NoV GI NoV GII

0.3 0.4

1.6 1.6

1.9 2.0

WWTP-4 NoV GI NoV GII

0.8 0.7

1.1 1.2

1.9 1.9

AS = activated sludge.

NoV GI and GII positive samples obtained by qPCR assay were amplified through nested PCR and the results obtained in these assays for the different treatment stages in the four WWTPs are presented in Table 6.

Table 6 Percentage (%) of positive samples obtained by nested PCR in different stages of the WWTPs. Stages

WWTP-1

WWTP-2

WWTP-3

WWTP-4

NoV GI (n positive sample in Nested PCR/n positive sample in qPCR) Raw sewage 90.9% (10/11) 91.6% (11/12) 83.3% (10/12) 83.3% (10/12) Secondary 83.3% (10/12) 71.4% (5/7) 0 (0/0) 37.5% (3/8) Reclaimed water 66.6% (2/3) 50.0% (2/4) 0 (0/0) 0 (0/0) NoV GII (n positive sample in Nested PCR/n positive sample in qPCR) Raw sewage 100% (12/12) 75.0% (9/12) 100% (12/12) 100% (11/11) Secondary 66.6% (8/12) 62.5% (5/8) 66.6% (2/3) 50.0% (4/8) Reclaimed water 57.1% (4/7) 25.0% (1/4) 0 (0/0) 0 (0/0)

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

Out of the total samples collected during the study (n = 141), 44.6% (63/141) were positive for NoV GI through the nested PCR protocol and 48.2% (68/141) positive for NoV GII (Table 6). NoV GI was genotyped in 42/63 (66.6%) of the sequenced samples. The results of the phylogenetic analysis of NoV GI are shown in Fig. 3. The following distribution of NoV GI genotypes was observed: NoV GI.2 (21/42) (52.3%), GI.5 (11/42) (26.1%), GI.3 subcluster a (8/42) (19.0%) and GI.3 subcluster b (1/42) (2.3%). A sample of NoV GI.2 (GenBank accession number: MF784762), one of NoV GI.3a (MF784761) and three of NoV GI.5 (MF784765, MF784769 and MF784787) were excluded from the phylogenetic tree because they had a shorter sequence length (≤294 nt).

433

The greatest diversity of genotypes was observed in the raw sewage samples, with a high prevalence of GI.2 genotype. NoV GI.5 and NoV GI.3a were detected in reclaimed water samples from WWTPs 1 and 2, respectively (Fig. 3). Out of the 68 positive samples for NoV GII, only 26 (38.2%) could be genotyped. Some samples, mainly those from secondary effluent and reclaimed water, could not be genotyped due to insufficient DNA or poor quality of sequences obtained after sequencing. A predominance of genotype GII.17 (13/26) (50.0%) was observed, followed by GII.4 Sydney 2012 variants (6/26) (23.0%), GII.6, GII.1 (3/ 26) (11.5%) and GII.5 (1/26) (3.8%) (Fig. 4).

Fig. 3. Phylogenetic tree obtained with NoV GI region C sequences (294 nt). Bootstrap values above 70% are indicated on the nodes. The sequences obtained in this study are named by the respective GenBank accession numbers, followed by the numbers designated in the laboratory, WWTP, sample type (RS = raw sewage - ■; TE = treated effluent - ●; RW = reclaimed water - ▲) and collection date (month and year). Reference sequences for each genotype are named by their GenBank accession numbers followed by the genotype.

434

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

Fig. 4. Phylogenetic tree obtained with NoV GII region C sequences (275 nt). Bootstrap values above 70% are indicated on the nodes. The sequences obtained in this study are named by the respective GenBank accession numbers, followed by the numbers designated in the laboratory, WWTP, sample type (RS = raw sewage - ■; TE = treated effluent - ●; RW = reclaimed water - ▲) and collection date (month and year). Reference sequences for each genotype are named by their GenBank accession numbers followed by the genotype.

NoV GII.4 Sydney 2012 variant was identified in a reclaimed water sample at WWTP-1, indicating the persistence of viral genomes to the filtration and chlorination disinfection process (Fig. 4).

4. Discussion 4.1. Detection rates, concentrations and efficiency of secondary and tertiary treatments in NoV removal Sewage is an important environmental source for studying the epidemiology of pathogens with fecal-oral transmission routes (Kazama et al., 2017; Zhou et al., 2016), especially when the differential diagnosis to isolate and identify the specific etiologic agent is not performed as a

routine procedure in medical and hospital care or when the diagnosis is only requested in outbreak situations in the community. Data about the occurrence of Norovirus GI in Brazil wastewaters are scarce. Studies conducted in Rio de Janeiro have shown a prevalence of NoV GII in these environmental samples, with absence or low detection rates of NoV GI (Fumian et al., 2013; Victoria et al., 2010). The present study showed that both NoV GI and GII are widely disseminated in the São Paulo wastewaters. Although, NoV GI was detected in 94% of sewage samples, the Virology Center of the Adolfo Lutz Institute (IAL), from São Paulo State Health's Secretary, has reported a low prevalence of this genotype among clinical samples from gastroenteritis outbreaks. Out of the 177 clinical samples evaluated in 2015 and 98 evaluated in 2016, NoV GI was present in only 4.8 and 3.0% of the cases, respectively; while NoV GII was responsible for 92.1% of gastroenteritis in 2015 and

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

97% in 2016 (personal communication). This data indicates that NoV GI is underreported in the population or it may be associated with a considerable proportion of asymptomatic cases. NoV GI could also cause less severe symptoms which do not lead to clinic visits or hospitalization. Similar levels of norovirus concentration circulating in São Paulo sewage samples (median loads of 5.2 to 5.4 log10 gc L−1 for NoV GI and 5.4 to 5.6 log10 gc L−1 for NoV GII) has been reported in Europe and Asia, where norovirus loads have been higher than in the USA and New Zealand (Campos et al., 2016; Eftim et al., 2017). The higher viral loads of NoV GII in raw wastewaters observed in this study have also been described in other geographical regions (Campos et al., 2016; Eftim et al., 2017; Kauppinen et al., 2014). NoV densities in wastewaters fluctuated slightly during the study period, with higher values concentrated in the winter months. In tropical countries, NoV outbreaks do not present a homogeneous pattern for seasonal variation, mainly because many countries do not have a great temperature variation range throughout the year. However, NoV infections have been shown to occur more frequently between June and October in South American countries with temperate clime, such as Argentina, Uruguay and Southern Brazil (Poló et al., 2016). Peaks of NoV GII concentrations have also been observed during the winter months in raw and secondary sewage samples from Rio de Janeiro city, southeastern Brazil (Victoria et al., 2010), similarly to the results observed in this study. Elevated NoV loads remained in the secondary effluents after activated sludge treatment (105−103 gc L−1). The removal efficiency in this treatment process, ranged from 0.3 to 0.8 log10 for NoV GI and 0.4 to 1.4 log10 for NoV GII, which are similar values to the decay rates observed in other studies (Montazeri et al., 2015; Sano et al., 2016). The activated sludge treatment generally had a more significant impact on Norovirus GII removal in relation to GI, similarly to other previously reported studies (Da Silva et al., 2007; Montazeri et al., 2015). Usually, average efficiencies of NoV GI and GII removal in activated sludge systems are expected between 1 and 1.5 log10 units (Francy et al., 2012; Sano et al., 2016). The tertiary treatment effluents composed by sand-anthracyte filters delivered 1.1 to 1.6 log10 NoV GI and 0.7 to 1.6 log10 NoV GII reductions. Kauppinen et al. (2014) analyzed different sand filter configurations for effluent treatment and reported removal efficiencies between 0.6 and 2.2 log10 for NoV GI and 0.8 and 3.5 log10 for NoV GII. Higher removal efficiencies were obtained for a composite configuration of sand and biotite as a filter medium (grain size 0–0.8 mm, 0–2 mm, respectively) (Kauppinen et al., 2014). The temperature interference on the sand filters performance was also analyzed by Kauppinen et al. (2014), which found lower virus removal efficiencies during the colder months. The increase in temperature favors the microbial biofilm formation on the filter surface, increasing the removal capacity. In the present study, the seasonal variation was not correlated with the performance of filtration processes in NoV removal due to a large proportion of left-censored data set. Nevertheless, it was possible to observe that most of the tertiary effluent samples positive for NoV were detected during winter and early spring, period in which NoV concentrations in raw and secondary sewage were significantly higher than those observed in other seasons. The tertiary treatment system from WWTP-2 composed of an additional coagulation/flocculation/sedimentation step and containing zeolites at the filter media, not delivered significant higher NoV reduction when compared to sand-anthracite filter systems from WWTPs-1 and 3. Lee et al. (2017) evaluated the reduction of indigenous viruses by ultrafiltration with and without coagulation (polyaluminum chloride – PACl)/sedimentation (CS) in a pilot scale water reclamation plant in Okinawa, Japan, and observed that CS increased the mean efficiency removal of MS2 phages, mainly acting on viral inactivation measured through plaque assays.

435

Norovirus genomes were not detected in reclaimed water effluents produced by WWTPs 3 and 4. The absence NoV genomes in WWTP-3 tertiary effluents could be related to the direct entry of chemical sewage (nitrocellulose industry) in the system, which could be contributing to viral inactivation during the treatment stages. In a previous study on the detection of Bacteroides fragilis bacteriophages (GB-124) at different stages in these WWTPs, it was also reported the lowest phage detection rate in the secondary and tertiary effluents from WWTP-3, suggesting viral inactivation during the treatment stages (Prado et al., 2018). NoV values below the detection limit were expected in the tertiary treatment from WWTP-4, since the MBR and RO are recognized as high performance systems for removing virus and other emerging pathogens and micropollutants (Francy et al., 2012; Sano et al., 2016; Yasui et al., 2017). Sano et al. (2016), in a review study about risk management of viral infectious diseases in wastewater reclamation and reuse, compiled representative data about the virus removal efficiencies in MBR and conventional activated sludge processes by a meta-analysis approach. The log10 reduction of NoV GII in MBR ranged from 0.96 to 4.16, with an overall mean of 3.35 (Sano et al., 2016). Despite the high efficiency of MBR for viral removal, NoV have been observed in effluents from these systems (Francy et al., 2012; Miura et al., 2015). Therefore, the combination of MBR with an even more sophisticated system, such as RO, would be an excellent alternative to produce reclaimed water for less restricted uses. Studies evaluating NoV removal efficiency in composite ultrafiltration systems followed by RO showed a 7.7 log10 efficiency in NoV removal (Yasui et al., 2017). It is important to emphasize that human enteric virus concentrations in tertiary effluent samples are frequently below the analytical detection limits, and therefore the log removal efficiency has been calculated by using the detection limit of these samples, making it difficult to compare removal efficiencies among viruses or studies (Miura et al., 2015). Although viral genomes were only detected in a small proportion of the reclaimed water samples from sand-anthracite filters, the confirmation of these viral genomes through nucleotide sequencing demonstrated their persistence to the disinfection processes. Other studies, including the analysis of virus indicators and infectivity tests, should be conducted to assess public health risks arising from different reuse practices in urban settings. 4.2. Epidemiology of circulating NoV in the environment In this study nine NoV genotypes were found based on direct amplicon sequencing, revealing the most prevalent genotypes circulating in the population. Other approaches, based on newer technologies, as next generation sequencing (NGS), could provide a higher viral diversity present in the community. Sequencing analyses of NoV GI samples revealed a high prevalence of the NoV GI.2, followed by GI.5 and GI.3 cluster a. Many of these sequences were closely related, which demonstrates the continuous circulation of some transmission lineages in the local population during the evaluated period. NoV GI.3 has also been frequently detected in sewage samples, according to environmental monitoring data collected in recent years, especially in Asian countries (Kazama et al., 2017; Zhou et al., 2016). Interestingly, Kazama et al. (2017) also found a relative high percentage of NoV GI.3 subcluster a strains in sewage samples and no cases were reported in clinical samples, demonstrating that epidemiological surveillance has rarely detected strains of NoV GI in the population, despite their significant presence in wastewater. According to De Graaf et al. (2017) NoV GI have a higher association with waterborne infections compared to GII viruses, probably due their higher stability in water and limited removal efficiency during sewage treatment. The relatively high concentrations of NoV GI in environmental samples could increase the risk of waterborne outbreaks in the São Paulo city. However up to 99% of local population is served by drinking

436

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437

water coverage and epidemiological surveillance data suggest that foodborne route or person-to-person contact seem to play a more important role as NoV transmission pathways in the outbreaks recorded in São Paulo (SMS/COVISA, 2016). Although the pandemic variant NoV GII.4 Sydney 2012 was identified in this study, this was the first report of GII.17 predominance in the Brazilian wastewaters. A newly emerged GII·P17–GII.17 genotype has recently become the predominant strain in some parts of Asia (De Graaf et al., 2015, 2016). Some recent studies have also reported the emergence of this new genotype through epidemiological and/or environmental surveillance data in other parts of the world (De Graaf et al., 2015; Mabasa et al., 2017), including Brazil (Andrade et al., 2017). There is still no evidence that this emerging genotype is more virulent than GII.4 genotype, however, based on the analysis of genome region encoding the VP1 protein, it was observed that the GII.17 genotype contains recombinant genotypes from other viruses enclosing sequences from the region named ORF-1 (De Graaf et al., 2015). The acquisition of a new sequence in the ORF-1 region could potentially result in an increase in replication efficiency of the virus and could partly explain the increased outbreaks of acute gastroenteritis caused by this new genotype (De Graaf et al., 2015). In Brazil this variant was recently identified and the 2014 World Cup event is pointed out as one of the possible entry points of this variant in the country (Andrade et al., 2017). NoV GII.4 strains were predominant from April to August 2015, while NoV GII.17 strains were predominantly detected from August 2015 to March 2016. Sequences of NoV GII.17 obtained in this study showed a high degree of homology among them (99–100% nt identity) and, they are closely related (100% nt identity) with several NoV GII.17 strains previously identified, with emphasis on GII. 17/Gaithersburg/US/ 2014 (KY424350), GII.17 Kawasaki (NS-405 - Hong Kong 2014 (KT326180)), NoV GII.17 Kawasaki 308, identified in Hong Kong in 2015 (KU561249) and NoV GII.17 recently associated with sporadic gastroenteritis cases in Russia in 2016 (KY210964). Although our analyzes were based on partial gene sequencing (region C), and does not allow us to affirm with certainty if the sequences belong to these novel NoV GII. P17/GII.17 variants, the public health services and surveillance systems need to be prepared for a case of a potential increase of this emerging viral genotype in a near future.

5. Conclusions Data obtained in this study are important given the scarcity of papers reporting the efficiency of granular media filtration and combined processes using MBR/RO systems for enteric virus removal. Although sandanthracite filters and chlorination are economically more viable as tertiary effluent treatment, MBR/RO systems are more suitable for the production of reclaimed water where the application requires higher quality criteria. Reclaimed water from such systems would be more appropriate for less restricted uses, mainly considering future reuse applications, such as indirect potable reuse. This was the first report on detection and characterization of NoV in reclaimed water from Brazil. The prevalence of genotype GII.17 in the environmental samples is a warning for a possible change in the epidemiological pattern of NoV GII strains circulating in the region. However, only data from epidemiological surveillance can corroborate this prediction over time. It was concluded that the environmental approach is useful in epidemiological studies and provides a rapid diagnosis of the norovirus strains circulating in the population.

Funding This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo - FAPESP [grant number: 2013/26586-1].

Acknowledgement The authors wish to acknowledge the staff of Environmental Company of the São Paulo State - CETESB, especially Carlos Jesus Brandão, Vivian Baltazar Rocha, Marcelo Oliveira, Fernando de Caires and Waldemar Rodrigues for conducting sampling and field analysis. We would also like to thank Vilma Marques Santos Bonanno (in memoriam) from Cetesb for the essential assistance with samples processing. Conflict of interest The authors declare that they have no conflicts of interest. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.07.301. References Amarasiri, M., Kitajima, M., Nguyen, T.H., Okabe, S., Sano, D., 2017. Bacteriophage removal efficiency as a validation and operational monitoring tool for virus reduction in wastewater reclamation: review. Water Res. 121, 258–269. https://doi.org/10.1016/ j.watres.2017.05.035. Andrade, J.S.R., Fumian, T.M., Leite, J.P.G., Assis, M.R., Bello, G., Mir, D., Miagostovich, M.P., 2017. Detection and molecular characterization of emergent GII.P17/GII.17 norovirus in Brazil, 2015. Infect. Genet. Evol. 51, 28–32. https://doi.org/10.1016/j. meegid.2017.03.011. APHA/AWWA/WEF, 2012. Standard Methods for the Examination of Water and Wastewater. 22 ed. American Public Health Association, Inc., Washington, DC. Bartsch, S.M., Lopman, B.A., Ozawa, S., Hall, A.J., Lee, B.Y., 2016. Global economic burden of norovirus gastroenteritis. PLoS One 11 (4), e0151219. Bates, D.M., Debroy, S., 2004. Linear mixed models and penalized least squares. J. Multivar. Anal. 91 (1), 1–17. https://doi.org/10.1016/j.jmva.2004.04.013. Brinkman, N.E., Haffler, T.D., Cashdollar, J.L., Rhodes, E.R., 2013. Evaluation of methods using celite to concentrate norovirus, adenovirus and enterovirus from wastewater. J. Virol. Methods 193, 140–146. https://doi.org/10.1016/j.jviromet.2013.05.014. Bursac, Z., Gauss, C.H., Williams, D.K., Hosmer, D.W., 2008. Purposeful selection of variables in logistic regression. Source Code Biol. Med. 3, 17. https://doi.org/10.1186/ 1751-0473-3-17. Campos, C.J.A., Avant, J., Lowther, J., Till, D., Lees, D.N., 2016. Human norovirus in untreated sewage and effluents from primary, secondary and tertiary treatment processes. Water Res. 103, 224–232. https://doi.org/10.1016/j.watres.2016.07.045. Chambers, J.M., Cleveland, W.S., Kleiner, B., Tukey, P.A., 1983. Graphical Methods for Data Analysis. Wadsworth International Group, Belmont, CA. Croghan, C.W., Egeghy, P.P., 2003. Methods of Dealing With Values Below the Limit of Detection Using SAS. US-EPA, Research Triangle Park, NC. Da Silva, A.K., Le Saux, J.-C., Parnaudeau, S., Pommepuy, M., Elimelech, M., Le Guyader, F.S., 2007. Evaluation of removal of noroviruses during wastewater treatment, using realtime reverse transcription-PCR: different behaviors of genogroups I and II. Appl. Environ. Microbiol. 73 (24), 7891–7897. https://doi.org/10.1128/AEM.01428-07. De Graaf, M., Beek, J.V., Vennema, H., Podkolzin, A.T., Hewitt, J., Bucardo, F., Templeton, K., Mans, J., Nordgren, J., Reuter, G., Rasmussen, L.D., Iritani, N., Chan, M.C., Martella, V., Ambert-Balay, K., Vinjé, J., White, P.A., Koopmans, M.P., 2015. Emergence of a novel GII.17 norovirus - end of the GII.4 era? Euro Surveill. 20 (26). http://www. eurosurveillance.org/ViewArticle.aspx?Articleid=21178 (pii=21178). De Graaf, M., van Beek, J., Koopmans, M.P.G., 2016. Human norovirus transmission and evolution in a changing world. Nat. Rev. Microbiol. https://doi.org/10.1038/ nrmicro.2016.48. De Graaf, M., Villabruna, N., Koopmans, M.P.G., 2017. Capturing norovirus transmission. Curr. Opin. Virol. 22, 64–70. https://doi.org/10.1016/j.coviro.2016.11.008. Eftim, S.E., Hong, T., Soller, J., Boehm, A., Warren, I., Ichida, A., Nappier, S.P., 2017. Occurrence of norovirus in raw sewage - a systematic literature review and metaanalysis. Water Res. 111, 366–374. https://doi.org/10.1016/j.watres.2017.01.017. Fioretti, J.M., Ferreira, M.S.R., Victoria, M., Vieira, C.B., Xavier, M.P.T.P., Leite, J.P.G., Miagostovich, M.P., 2011. Genetic diversity of noroviruses in Brazil. Mem. Inst. Oswaldo Cruz 106 (8), 942–947. Francy, D.S., Stelzer, E.A., Bushon, R.N., Brady, A.M.G., Williston, A.G., Riddell, K.R., Borchardt, M.A., Spencer, S.K., Gellner, T.M., 2012. Comparative effectiveness of membrane bioreactors, conventional secondary treatment, and chlorine and UV disinfection to remove microorganisms from municipal wastewaters. Water Res. 46, 4164–4178. https://doi.org/10.1016/j.watres.2012.04.044. Fumian, T.M., Leite, J.P.G., Castello, A.A., Gaggero, A., de Caillou, M.S.L., Miagostovich, M.P., 2010. Detection of rotavirus A in sewage samples using multiplex qPCR and an evaluation of the ultracentrifugation and adsorption-elution methods for virus concentration. J. Virol. Methods 170, 42–46. https://doi.org/10.1016/j.jviromet.2010.08.017. Fumian, T.M., Vieira, C.B., Leite, J.P.G., Miagostovich, M.P., 2013. Assessment of burden of virus agents in an urban sewage treatment plant in Rio de Janeiro, Brazil. J. Water Health 11 (1), 110–119. https://doi.org/10.2166/wh.2012.123.

T. Prado et al. / Science of the Total Environment 646 (2019) 427–437 Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98. Kauppinen, A., Martikainen, K., Matikka, V., Veijalainen, A.-M., Pitkanen, T., HeinonenTanski, H., Miettinen, I.T., 2014. Sand filters for removal of microbes and nutrients from wastewater during a one-year pilot study in a cold temperate climate. J. Environ. Manag. 133, 206–213. https://doi.org/10.1016/j.jenvman.2013.12.008. Kazama, S., Miura, T., Masago, Y., Konta, Y., Tohma, K., Manaka, T., Liu, X., Nakayama, D., Tanno, T., Saito, M., Oshitani, H., Omura, T., 2017. Environmental surveillance of norovirus genogroups I and II for sensitive detection of epidemic variants. Appl. Environ. Microbiol. 83 (9), e03406-16. https://doi.org/10.1128/AEM.03406-16. Kirk, M.D., Pires, S.M., Black, R.E., Caipo, M., Crump, J.A., Devleesschauwer, B., Dopfer, D., Fazil, A., Fischer-Walker, C.L., Hald, T., Hall, A.J., Keddy, K.H., Lake, R.J., Lanata, C.F., Torgerson, P.R., Havelaar, A.H., Angulo, F.J., 2015. World Health Organization estimates of the global and regional disease burden of 22 foodborne bacterial, protozoal, and viral diseases, 2010: a data synthesis. PLoS Med. 12 (12), e1001921. https://doi. org/10.1371/journal.pmed.1001921. Kitajima, M., Oka, T., Haramoto, E., Takeda, N., Katayama, K., Katayama, H., 2010. Seasonal distribution and genetic diversity of genogroups I, II, and IV noroviruses in the Tamagawa River. Environ. Sci. Technol. 44, 7116–7122. https://doi.org/10.1021/ es100346a. Kitajima, M., Iker, B.C., Pepper, I.L., Gerba, C.P., 2014. Relative abundance and treatment reduction of viruses during wastewater treatment processes - identification of potential viral indicators. Sci. Total Environ. 488-489, 290–296. https://doi.org/10.1016/j. scitotenv.2014.04.087. Kroneman, A., Vennema, H., Deforche, K., Avoort, H.V., Peñaranda, S., Oberste, M.S., Vinjé, J., 2011. Koopmans An automated genotyping tool for enteroviruses and noroviruses. J. Clin. Virol. 51 (2), 121–125. Lee, S., Hata, A., Yamashita, N., Tanaka, H., 2017. Evaluation of virus reduction by ultrafiltration with coagulation-sedimentation in water reclamation. Food Environ. Virol. https://doi.org/10.1007/s12560-017-9301-9. Lopman, B., Gastañaduy, P., Park, G.W., Hall, A.J., Parashar, U.D., Vinjé, J., 2012. Environmental transmission of norovirus gastroenteritis. Curr. Opin. Virol. 2, 96–102. Mabasa, V.V., Meno, K.D., Taylor, M.B., Mans, J., 2017. Environmental surveillance for noroviruses in selected south African wastewaters 2015–2016: emergence of the novel GII.17. Food Environ. Virol. https://doi.org/10.1007/s12560-017-9316-2. Miura, T., Okabe, S., Nakahara, Y., Sano, D., 2015. Removal properties of human enteric viruses in a pilot-scale membrane bioreactor (MBR) process. Water Res. 75, 282–291. https://doi.org/10.1016/j.watres.2015.02.046. Montazeri, N., Goettert, D., Achberger, E.C., Johnson, C.N., Prinyalwatkul, W., Janes, M.E., 2015. Pathogenic enteric viruses and microbial indicators during secondary treatment of municipal wastewater. Appl. Environ. Microbiol. 81, 6436–6445. https:// doi.org/10.1128/AEM.01218-15. Municipal Health Secretary - SMS. Coordination of Health Surveillance - COVISA, 2016. Epidemiological surveillance of waterborne and foodborne diseases. Available on:. http://prefeitura.sp.gov.br/covisa, Accessed date: 4 October 2018. Nelder, J.A., Wedderburn, R.W.M., 1972. Generalized linear models. J. R. Stat. Soc. Ser. A 370–384. O'Ryan, M., Riera-Montes, M., Lopman, B., 2017. Norovirus in Latin America: systematic review and meta-analysis. Pediatr. Infect. Dis. J. 36 (2), 127–134. https://doi.org/ 10.1097/INF.0000000000001369. Otto, T.D., Vasconcellos, E.A., Gomes, L.H.F., Moreira, A.S., Degrave, W.M., Mendonça-Lima, L., Alves-Ferreira, M., 2008. ChromaPipe: a pipeline for analysis, quality control and management for a DNA sequencing facility. Genet. Mol. Res. 7, 861–871.

437

Poló, T.S., Peiró, J.R., Mendes, L.C.N., Ludwig, L.F., de Oliveira-Filho, E.F., Bucardo, F., Huynen, P., Melin, P., Thiry, E., Mauroy, A., 2016. Human norovirus infection in Latin America. J. Clin. Virol. 78, 111–119. https://doi.org/10.1016/j.jcv.2016.03.016. Prado, T., Bruni, A.C.B., Barbosa, M.R.F., Bonanno, V.M.S., Garcia, S.C., Sato, M.I.Z., 2018. Distribution of human fecal marker GB-124 bacteriophages in urban sewage and reclaimed water of São Paulo City, Brazil. J. Water Health 16 (2), 289–299. https:// doi.org/10.2166/wh.2017.011. Rajal, V.B., McSwain, B.S., Thompson, D.E., Leutenegger, C.M., Kildare, B.J., Wuertz, S., 2007. Validation of hollow fiber ultrafiltration and real-time PCR using bacteriophage PP7 as surrogate for the quantification of viruses from water samples. Water Res. 41, 1411–1422. https://doi.org/10.1016/j.watres.2006.12.034. Resolution SES/SMA/SSRH No. 1, 2017. “Disciplina o reúso direto não potável de água, para fins urbanos, proveniente de Estações de Tratamento de Esgoto Sanitário e dá providências correlatas”. São Paulo's State Government. DOE 29-06-2017 Section I page 41/42. Published [June 28, 2017]. Rhodes, E.R., Hamilton, D.W., See, M.J., Wymer, L., 2011. Evaluation of hollow-fiber ultrafiltration primary concentration of pathogens and secondary concentration of viruses from water. J. Virol. Methods 176, 38–45. https://doi.org/10.1016/j. jviromet.2011.05.031. Sano, D., Amarasiri, M., Hata, A., Watanabe, T., Katayama, H., 2016. Risk management of viral infectious diseases in wastewater reclamation and reuse: review. Environ. Int. 91, 220–229. https://doi.org/10.1016/j.envint.2016.03.001. Sidhu, J.P.S., Sena, K., Hodgers, L., Palmer, A., Toze, S., 2017. Comparative enteric viruses and coliphage removal during treatment processes in a sub-tropical environment. Sci. Total Environ. 616-617, 669–677. https://doi.org/10.1016/j. scitotenv.2017.10.265. Smith, C.M., Hill, V.R., 2009. Dead-end-hollow-fiber ultrafiltration for recovery of diverse microbes from water. Appl. Environ. Microbiol. 75 (16), 5284–5289. https://doi.org/ 10.1128/AEM.00456-09. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28, 2731–2739. https://doi.org/ 10.1093/molbev/msr121. USEPA - Environmental Protection Agency of United States, 2014. Method 1615 - Measurement of Enterovirus and Norovirus. Occurrence in Water by Culture and RTqPCR. Version 1.3, Revised September 2014. Victoria, M., Guimarães, F.R., Fumian, T.M., Ferreira, F.F.M., Vieira, C.B., Shubo, T., Leite, J.P.G., Miagostovich, M.P., 2010. One year monitoring of norovirus in a sewage treatment plant in Rio de Janeiro, Brazil. J. Water Health 8 (1), 158–165. https://doi.org/ 10.2166/wh.2009.012. Vinjé, J., 2015. Advances in laboratory methods for detection and typing of norovirus. Mini review. J. Clin. Microbiol. 53 (2), 373–381. https://doi.org/10.1128/JCM.0153514. Yasui, N., Suwa, M., Minamiyama, M., 2017. Infectious risk assessment of reclaimed water by UF membrane treatment process focusing attention on norovirus. Water Sci. Technol. https://doi.org/10.2166/ws.2017.114. Zhou, N., Lin, X., Wang, S., Tao, Z., Xiong, P., Wang, H., Liu, Y., Song, Y., Xu, A., 2016. Molecular epidemiology of GI and GII noroviruses in sewage: 1-year surveillance in eastern China. J. Appl. Microbiol. 121 (4), 1172–1179. https://doi.org/10.1111/jam.13218.