Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Detection and Characterization of Hepatitis E Virus Genotype 3 in Wastewater and Urban Surface Waters in Germany

Detection and Characterization of Hepatitis E Virus Genotype 3 in Wastewater and Urban Surface... In highly populated areas, environmental surveillance of wastewater and surface waters is a key factor to control the circu- lation of viruses and risks for public health. Hepatitis E virus (HEV) genotype 3 is considered as an emerging pathogen in industrialized countries. Therefore, this study was carried out to determine the prevalence of HEV in environmental waters in urban and suburban regions in Germany. HEV was monitored in water samples using quantitative RT-PCR (RT-qPCR) and nested RT-PCR without or with virus concentration via polyethylene glycol precipitation or ultracentrifugation. By RT- qPCR, 84–100% of influent samples of wastewater treatment plants were positive for HEV RNA. Genotypes HEV-3c and 3f were identified in wastewater, with HEV-3c being the most prevalent genotype. These data correlate with subtypes identified earlier in patients from the same area. Comparison of wastewater influent and effluent samples revealed a reduction of HEV RNA of about 1 log during passage through wastewater treatment plants. In addition, combined sewer overflows (CSOs) after heavy rainfalls were shown to release HEV RNA into surface waters. About 75% of urban river samples taken during these CSO events were positive for HEV RNA by RT-qPCR. In contrast, under normal weather conditions, only around 30% of river samples and 15% of samples from a bathing water located at an urban river were positive for HEV. Median concentrations of HEV RNA of all tested samples at this bathing water were below the limit of detection. Keywords Hepatitis E virus · Monitoring · Genotyping · Wastewater · Surface water · Combined sewer overflow Introduction Genotypes 3 and 4 are zoonotic and infect mainly humans, swine and wild boars (Pavio et al. 2017). Whereas genotype Hepatitis E virus (HEV) is the causative agent of acute 4 is mainly restricted to Asia, in most industrialized coun- and chronic hepatitis in humans worldwide. A severe dis- tries genotype 3 is predominant (Clemente-Casares et al. ease progression is possible with mortality rates around 2003; Meng 2010; Dalton et al. 2014). 1% (Pérez-Gracia et al. 2015). However, among pregnant In Germany seroprevalences for HEV-specific antibodies women infected with HEV genotype 1, a higher incidence of about 1% in children (Krumbholz et al. 2014) and about and severity was observed with mortality rates up to 30% 15% among adults (Faber et al. 2018a) were reported. A (Clemente-Casares et al. 2016). continuous increase in the number of notified hepatitis E HEV is classified into four main human-pathogenic gen- cases was recorded in Germany during the last years, most otypes within the Hepeviridae family. Genotypes 1 and 2 likely due to increased awareness. In 2018 about 3400 new infect only humans and are endemic in developing countries. hepatitis E cases were reported to the Robert Koch Institute (RKI 2019). HEV is transmitted mainly through meat products of * Hans-Christoph Selinka infected animals and faecally contaminated water. Trans- hans-christoph.selinka@uba.de mission to humans through contaminated water is known for Section II 1.4 Microbiological Risks, German Environment genotypes HEV-1 and HEV-2, mainly in developing coun- Agency (UBA), Corrensplatz 1, 14195 Berlin, Germany tries (Fenaux et al. 2019). Berliner Wasserbetriebe (BWB), Cicerostr. 24, 10709 Berlin, Industrialization of a country decreases HEV risk related Germany to HEV-1, but increases that related to HEV-3 and HEV-4, German Federal Institute for Risk Assessment (BfR), as observed in China (Sridhar et al. 2015). Max-Dohrn-Straße 8-10, 10589 Berlin, Germany Vol.:(0123456789) 1 3 138 Food and Environmental Virology (2020) 12:137–147 So far, the role of water in the transmission of zoonotic taken by the Berlin Centre of Competence for Water during HEV-3 has only been suspected (Fenaux et al. 2019). A a sampling campaign after heavy rainfall events in 2016. recent study identified the occupational contact with waste- Additional samples were drawn and analysed from a bathing water as a risk factor associated with autochthonous hepatitis water located at river 2 (river 2/bathing water) in the years E in Germany, supporting that waterborne transmission of 2018 and 2019. Water samples were processed directly after HEV-3 is possible (Faber et al. 2018b). sampling or stored at − 80 °C until further processing. In developed countries human and animal hosts of HEV-3 may contaminate wastewater through their faeces. HEV par- Sample Concentration ticles can reach the environment and potentially contaminate surface waters. Thus, surface waters could be a source of In environmental samples, human-pathogenic viruses are HEV contamination for animals and humans (Fenaux et al. mostly present in low or very low concentrations and have 2019). to be further concentrated for analyses. In our study we used Increasing HEV prevalence in industrialized countries is ultracentrifugation (U) and polyethylene glycol (PEG) pre- known since 1998. In Spain, HEV detection in urban sewage cipitation (Fig. 1). samples was reported (Pina et al. 1998), followed by reports Ultracentrifugation was performed according to a pre- from the Netherlands (Rutjes et al. 2009), Italy (La Rosa viously described method (La Rosa et al. 2007). In brief, et al. 2010) and other countries. Most common detection 180 ml supernatant after initial centrifugation at 3000×g for methods are nested reverse transcription (RT) polymerase 10 min was pelleted by 2 h centrifugation at 160,000×g with chain reaction (PCR) or quantitative RT-PCR (RT-qPCR) a 45Ti rotor in an Optima L-100 K ultracentrifuge (Beck- with or without prior virus concentration steps. In recent mann, Germany) and resuspended in 5 ml PBS for nucleic years, Italy, Norway and the UK have reported first investi- acid extraction. gations for a HEV surveillance in sewage (Idolo et al. 2013; For virus concentration by PEG precipitation (Manor Myrmel et al. 2015; Smith et al. 2016a; Alfonsi et al. 2018). et al. 2007), PEG 6000 (80 g) and NaCl (17.5 g) were added To the best of our knowledge there are no available stud- to 1 l water samples, mixed for 1 h and stored overnight ies on the presence of HEV in environmental waters in Ger- at 4 °C. Subsequently, precipitates containing the viruses many. Therefore, this study was carried out to (1) investigate were collected after 1 h centrifugation at 12,200×g. Pel- the HEV prevalence in environmental water samples, (2) lets were resuspended in 15 ml PBS, 15 ml chloroform (to to compare HEV concentration methods, and (3) to geno- destroy bacteria) and 150 µl Tween-80. After centrifuga- type detected HEV strains. Wastewater influent and efflu- tion for 15 min at 1400×g, the top layer was saved and the ent samples of urban and suburban wastewater treatment lower chloroform layer was removed. Remaining pellets plants (WWTPs), surface waters from two rivers including were resuspended in 0.05 mol/l glycine pH 7.2 with 3% beef a bathing water and conditions of combined sewer overflows extract. Centrifugation was repeated and both supernatants (CSO) were investigated. For comparison of virus concen- tration techniques, PEG-precipitated samples, samples sub- jected to ultracentrifugation and samples without further virus concentrations were tested simultaneously. Genotyping was performed for further characterization of the detected HEV strains. Material and Methods Sampling Samples of wastewater influents (after coarse grid removal) and wastewater effluents (secondary effluents, before UV treatment) of WWTPs were collected in the years 2014–2019 from central urban (WWTP 1) and suburban (WWTPs 2–4) areas of the cities of Berlin and Munich, Germany. Surface water samples were taken in the years 2016–2019 from two urban rivers at normal weather conditions (river 1 and river 2), as well as after heavy rainfall events with com- Fig. 1 Flow chart of methods applied on wastewater and river water bined sewer overflows (river 1/CSO). CSO samples were samples for HEV RNA detection 1 3 Food and Environmental Virology (2020) 12:137–147 139 were combined to a final volume of about 20 ml. Exact vol- not detected, the LOD concentration was used for further umes were noted for calculating virus concentrations in the calculations. With 200 µl of viral nucleic acids eluted by the original samples. NucliSENS® easyMAG® method from 5 ml of direct water samples, the LOD was 200 copies/100 ml. Using nucleic Nucleic Acid Extraction acids from water samples concentrated by ultracentrifuga- tion, an LOD of 6 copies/100 ml was achieved and the LOD Nucleic acid extraction was performed with 5 ml volumes of PEG-precipitated samples was four copies per 100 ml. of concentrated samplesor 5 ml volumes of samples with- The limit of quantification (LOQ) was set ten times higher out further virus concentration steps (direct samples). The than the LOD of each method. NucliSENS® easyMAG® (bioMérieux, Germany) method allows simultaneous extraction of DNA and RNA with HEV‑Specific Nested RT‑PCR same efficiencies. To assess the extraction efficacy, a sam- ple spiked with human adenovirus 2 with a defined con- Primers for nested RT-PCR, which amplify a 332 bp product centration was included for each set of samples subjected from the HEV open reading frame 1 (ORF1) were designed to the nucleic acid extraction procedure. The method was by Johne et al. (2010). RNA from the HEV isolate 47832c used according to the manufacturer’s protocol, with slight (Johne et al. 2014) was used as positive control. For the modifications. Centrifugation was performed at 6000×g for first RT-PCR with a total reaction volume of 25 µl, 5 µl of 5 min after lysis buffer incubation to eliminate large disturb- template was amplified using the OneStep Ahead RT-PCR ing particles present in turbid water samples. In addition, Kit (QIAGEN, Germany). Cycling profile included the fol- purified nucleic acids were eluted two times in 100 µl elution lowing settings: 10 min at 50 °C, 5 min at 95 °C, 40 cycles buffer resulting in a final volume of 200 µl to allow analyses of 10 s at 95 °C, 10 s at 55 °C, 10 s at 72 °C and 2 min at of several qPCR reactions. 72 °C. The second nested PCR was performed with 2 µl template from the first RT-PCR. The Taq DNA Polymerase HEV‑Specific Quantitative Real‑Time RT‑PCR Kit (QIAGEN, Germany) was used according to protocol in a total reaction volume of 50 µl. Primer concentrations were Primers designed by Jothikumar et al. (2006) were used for 0.3 µM and cycling conditions were the following: 3 min quantitative RT-PCR. Probes were used either as described at 94 °C, 35 cycles of 45 s at 94 °C, 45 s at 60 °C, 1 min at by Jothikumar et al. (2006) or in a modified version (Gar - 72 °C and 10 min at 72 °C. son et al. 2012) using the quencher MGB (Minor groove PCR fragments were separated by gel electrophoresis on binder) to reduce the risk of false negative real-time RT-PCR 1.5% agarose gels in 1 × TBE buffer with 10 µl of 10,000×g results. RT-qPCR was performed in a volume of 25 µl using GelRed staining (Biotium, Germany) per 100 ml agarose the QuantiTect Probe RT-PCR Kit (QIAGEN, Germany) solution. Loading buffer (Thermo Scientific, Germany) was with a probe concentration of 0.2 µM. The following cycle mixed with the PCR products and gels were run for 50 min conditions were applied: 30 min at 50 °C, 15 min at 95 °C at 90 V. Low range DNA ladder (5 µl) was used as a size and 45 cycles with 15 s at 94 °C and 1 min at 56 °C. Each marker (Thermo Scientific, Germany). reaction mix contained 10 µl of undiluted or 1:10 diluted templates (4 reactions per sample) to detect putative inhibi- DNA Sequencing and Nucleotide Sequence Analyses tion of the RT-qPCR reaction. Copy numbers are calculated based on all reactions, if undiluted and diluted samples cor- Bands of the expected length (332 bp) were excised and respond. In the case of partial inhibition in the undiluted purified according to the protocol from innuPREP DOU- samples, copy numbers of the diluted samples were chosen BLEpure Kit (Analytik Jena, Germany). The cDNA was for quantification. eluted twice in 30 µl elution buffer, combined and sequenced Double-stranded DNA Gene Strands (Eurofins Genom- by Eurofins (Germany). ics, Sweden) containing the specific amplification sequence All HEV sequences determined in this study were were applied as quantitative HEV standards in concentra- submitted to NCBI GenBank under accession numbers 6 1 tions from 10 HEV copies/10 µl to 10 HEV copies/10 µl MT087290 to MT087304. Sequence alignments and to generate a standard curve for determination of virus copy phylogenetic trees were constructed with Molecular numbers in the samples. Standard deviations of samples Evolutionary Genetics Analysis Version 7.0 (MEGA 7) during the 1-year surveillance were calculated from two to software (Kumar et  al. 2016). The MUSCLE program seven monthly samples. The calculation of the limit of detec- was used for multiple sequence alignment and maxi- tion (LOD) of HEV RNA was based on duplicates of 10 µl mum likelihood as statistic method based on the Kimura nucleic acid templates per RT-qPCR reaction with at least 1 2-parameter model (Kimura, 1980). The phylogenetic HEV copy to be detected in the duplicate. If HEV RNA was trees were validated by replicating with 1000 bootstraps. 1 3 140 Food and Environmental Virology (2020) 12:137–147 Obtained HEV sequences were aligned to 41 HEV-sub- Results type reference sequences (or a subset of 19 genotype 3 reference subtype sequences), as recommended by Smith Monitoring of HEV in Environmental Water Samples et al. (2016b). In addition, sequences were aligned to the by RT‑qPCR HEV-3c positive control (isolate 47832c from Johne et al. 2014) and 17 sequences from HEV infected patients from Surface waters as well as influent and effluent wastewater the Charité Hospital in Berlin (Wang et al. 2018a). samples of WWTPs from urban and suburban areas were monitored for HEV by RT-qPCR. Four urban and suburban WWTPs, differing in their catchment areas and cleaning Statistical Analyses capacities, were investigated (Table  1). Of 111 wastewa- ter influent samples collected in the urban WWTP 1, 84% Statistical analyses were performed with Microsoft Excel. were positive for HEV RNA with a median concentration of As quantification data are not normally distributed but 3 × 10 copies/100 ml. The median concentration of all 111 ordinally scaled, the Mann–Whitney U test was used to tested samples was 2 × 10 copies/100 ml. determine the statistical significance at a 95% confidence HEV RNA was also detected in 26 out of 83 wastewater level. This test was carried out to evaluate statistical dif- effluent samples (31%) of WWTP 1 with a median concen- ferences between monthly virus concentrations of WWTP tration of 1 × 10 copies/100 ml in positive samples. How- inf luent samples and WWTP eff luent samples (Fig. 2) as ever, the median concentration of all 83 tested effluent sam- well as between different virus concentrations methods ples was below the limit of detection (LOD). (Fig. 3). In three suburban WWTPs (WWTP 2–4), HEV RNA was detected in 86–100% of the influent samples with median concentrations of positive samples in the range of 2 × 10 copies/100  ml–1 × 10 copies/100 ml. Effluent samples of Fig. 2 Comparison of HEV RNA concentrations in monthly influent samples (I) and efflu- ent samples (E) of WWTP 1, analysed by RT-qPCR without virus concentration steps. Black and grey bars represent meas- ured HEV concentrations above the LOD (open bars). LOD is the limit of detection with 200 copies/100 ml Fig. 3 Concentration of HEV RNA in monthly influent sam- ples of an urban WWTP over a period of one year. For com- parison of sampling methods direct samples (D) and samples concentrated by ultracentrifuga- tion (U) and PEG precipitation (P) were analysed by RT-qPCR. Black bars represent measured HEV concentrations above the LOD (open bars), which differ in each method 1 3 Food and Environmental Virology (2020) 12:137–147 141 Table 1 Detection of HEV RNA in WWTP influent and effluent samples by RT-qPCR Cleaning WWTP Influent samples WWTP Effluent samples capacity 3 Tested (n) Positive Median positive Median all Tested (n) Positive Median positive Median all [m /day] [n/ (%)] samples* samples* [n/ (%)] samples* samples* 3 3 3 WWTP 1 257.000 111 93 (84%) 3 × 10 2 × 10 83 26 (31%) 1 × 10 < LOD 3 3 2 WWTP 2 5.500 10 9 (90%) 2 × 10 1 × 10 2 1 (50%) 8 × 10 4 × 102 3 3 2 WWTP 3 119.000 7 6 (86%) 4 × 10 3 × 10 3 3 (100%) 4 × 10 4 × 102 4 4 WWTP 4 40.000 6 6 (100%) 1 × 10 1 × 10 nt nt nt nt WWTP wastewater treatment plant, LOD limit of detection, nt not tested *[HEV copies/100 ml] was 3 × 10 copies/100 ml, but the median concentration of Table 2 Detection of HEV RNA in surface water samples by RT- qPCR all 55 tested samples was below the LOD. Surface water samples One‑year HEV Surveillance of a Wastewater Tested Positive Median positive Median all Treatment Plant (n) [n/(%)] samples* samples* River 1 21 7 (33%) 6 × 10 < LOD To investigate if the high variability in the concentrations of 3 3 River 1/CSO 16 12 (75%) 2 × 10 2 × 10 HEV in influent and effluent samples was affected by envi- River 2 69 21 (30%) 9 × 102 < LOD ronmental or seasonal influences, the central urban WWTP River 2/BW 55 8 (15%) 3 × 102 < LOD 1 was surveilled during a complete cycle of a year (Fig. 2). In WWTPs, virus concentrations are sufficiently high LOD limit of detection, BW bathing water, CSO combined sewer to be detected in small volumes without further virus con- overflow centration steps. Therefore, direct samples were measured * [HEV copies/100 ml] from March 2018 to February 2019 in influent and effluent samples of WWTP 1. HEV RNA was detected in 11 out of 12 monthly influent samples and in eight out of 12 effluent suburban WWTPs were positive for HEV RNA at rates of samples based on the LOD of 200 copies/100 ml. 50% (WWTP 2) and 100% (WWTP 3). Median concentra- Influent and effluent samples were taken the same day tions of these positive samples were 8 × 10 copies/100 ml without considering the passage time of wastewater treat- and 4 × 10 copies/100 ml, respectively. Median concen- ment. Several samples were collected each month. The mean trations of all tested effluent samples of these suburban concentration of all monthly measured samples is shown in WWTPs were 4 × 10 copies/100 ml. the figure for each month and was used to compare influent The results for HEV monitoring of surface waters are and effluent samples of WWTP 1. shown in Table 2. About 30% of 90 tested samples of two The mean value of calculated HEV RNA concentrations urban rivers under normal weather conditions (river 1 and of 12 monthly influent samples over the surveilled year was river 2) were positive for HEV RNA with median concen- 2 × 10 genome copies/100 ml. Effluent samples resulted in 2 2 3 trations of 6 × 10 copies/100 ml and 9 × 10 copies/100 ml, a mean of 2 × 10 copies/100 ml over this one-year period. respectively. Although effluents of WWTP 1 are released The average HEV RNA reduction during the passage of the into river 2 about 3 miles upstream of the sampling site, WWTP was about 1 log , comparing influent and effluent median concentrations of all river samples were below the samples above the LOD. Moreover, HEV concentrations of LOD. However, after heavy rainfall events, causing com- influent samples are significantly higher than from effluent bined sewer overflows (CSOs) upstream into river 1 (river 1/ samples (Mann–Whitney U test, p < 0.05). With a limit of CSO), 75% of the samples were positive for HEV RNA with quantification (LOQ) set to tenfold LOD, 10 of 12 influent a median concentration of 2 × 10 copies/100 ml. samples and only 2 effluent samples were positive for HEV In a bathing water located at the urban river 2 (river 2/ RNA, demonstrating the clearing effect of at least 1 log in BW) downstream the first sampling site of river 2, only eight the wastewater treatment plant. During the surveilled year, out of 55 samples (15%) were positive for HEV. The median no obvious seasonal pattern of HEV occurrence in wastewa- concentration of eight positive samples of this bathing water ter samples was observed. 1 3 142 Food and Environmental Virology (2020) 12:137–147 1 3 Food and Environmental Virology (2020) 12:137–147 143 ◂Fig. 4 Characterization of HEV strains from wastewater samples by 94 wastewater effluent samples and 57 river water samples gel electrophoresis and sequencing. a Exemplary agarose gel with showed a clear 332 bp band in the nested RT-PCR suitable HEV positive samples (332 bp fragments) from two WWTP influent for sequencing (data not shown). samples. b Maximum likelihood phylogenetic consensus tree of HEV Genotyping and subtyping were performed by sequence strains detected in urban wastewaters. Numbers at the nodes repre- sent bootstrap values > 60. Scale bar indicates the genetic distance alignments with reference strains followed by phylogenetic (nucleotide substitutions per site). Identified HEV sequences detected analyses (Fig.  4b). All identified sequences belonged to in wastewater samples are marked with a black dot. Names consist HEV genotype 3. Therefore, only reference sequences of of accession numbers, places, months, years of sampling and prepa- genotype 3 are shown. Bootstrap values > 60 are reported. ration methods (D: direct sample, U: ultracentrifugation, PEG: poly- ethylene glycol precipitation). Sequences from HEV infected patients HEV genotype 3c was the most prevalent subtype detected are labelled with open dots. HEV sequences were aligned to 41 in10 wastewater influent samples. Two wastewater samples HEV-subtype reference sequences denoted by accession number, sub- were identified as HEV genotype 3f. For three other samples genotype and source of first detection. Since all identified sequences no subtypes were classified. Of these 15 genotyped HEV belonged to genotype HEV-3, only sequences of this genotype are shown. Three rabbit HEV-3 sequences were used as outgroup strains from wastewater samples, ten were obtained from samples concentrated by ultracentrifugation, three from sam- ples prepared by PEG precipitation and two from samples Comparison of HEV Concentration Methods without further virus concentration steps. Moreover, the HEV genotypes identified in samples of urban and subur - To investigate if sample preparation methods have an impact ban WWTPs from the years 2016–2019 were compared to sequences of HEV infected patients from the same area from on the detection rate and the measured HEV RNA concen- trations, direct samples and samples concentrated by ultra- 2009–2016 (Wang et al. 2018a). As seen in the phylogenetic tree (Fig. 4), most of the wastewater and patient sequences centrifugation and polyethylene glycol precipitation were compared over a cycle of one year using influent samples of cluster in subtype 3c or 3f. WWTP 1 (Fig. 3). Each of these three methods has a different limit of Discussion detection, namely 4 copies/100 ml or 6 copies/100 ml for PEG precipitation and ultracentrifugation, respectively, or This study presents a quantitative surveillance and geno- 200 copies/100 ml for direct samples. In direct wastewater influent samples, HEV RNA was detected in 11 out of 12 typing of HEV strains in urban and suburban wastewater influent and effluent samples as well as in surface waters. monthly samples. In samples concentrated by ultracentrifu- gation, HEV RNA was found each month. Calculated HEV The zoonotic genotype 3 of HEV is autochthonous in many industrialized countries (Clemente-Casares et  al. concentrations in direct samples and samples concentrated by ultracentrifugation were in the similar range, in contrast 2003; Meng 2010; Dalton et al. 2014). Besides foodborne transmission of this genotype, environmental transmission to PEG-processed samples, which resulted in lower HEV RNA concentrations and lower detection rates. Using the pathways have also been proposed. In the present study, a wide distribution of HEV RNA in environmental waters in PEG method, viruses were detected only three times during this surveillance year and thus they were clearly signic fi antly Germany was identified, which may pose a risk of environ- mental transmission of HEV. However, HEV RNA detected different from direct samples and samples concentrated by ultracentrifugation (Mann–Whitney U test, p < 0.05). by PCR methods does not necessarily represent intact and infective virus particles. Genotyping of HEV from Environmental Water The highest detection rates (84–100%) of HEV RNA by quantitative PCR were found in wastewater influent sam- Samples ples, with a detection rate of 84% in WWTP 1 (Table 1). In contrast, only 31% of the effluent samples of WWTP 1 were To characterize the detected HEV strains in more detail, sequencings were carried out to identify HEV genotypes and positive for HEV RNA, demonstrating a cleaning effect of the WWTP with regard to HEV. In accordance with this subgenotypes in urban and suburban water samples (Fig. 4). After performing nested RT-PCR with HEV-specific finding, quantitative data on all tested samples of WWTPs indicate an HEV RNA reduction of about 1 log during primers, wastewater influent samples of two different WWTPs (WWTP 1 and WWTP 3) clearly showed the char- treatment. This result was validated by the HEV surveil- lance of WWTP 1 over a complete one-year period, showing acteristic 332 bp fragments. An exemplary agarose gel with amplified HEV nested RT-PCR products from HEV ORF1 an average decrease from 2 × 10 genome copies/100 ml in influent samples to 2 × 10 copies/100 ml in effluent samples is shown in Fig. 4a. Out of 173 tested wastewater influent samples, 15 samples (9%) displayed a clear band on the gel (Fig. 2). These effluent samples were taken before further UV treatment in the WWTPs. However, in summer, HEV and fragments were subjected to sequencing. None of the 1 3 144 Food and Environmental Virology (2020) 12:137–147 RNA reduction during wastewater treatment is expected to Detection and quantification of HEV RNA in environ- be higher, since WWTP 1 is run during the bathing season mental water samples is challenging. If low virus concentra- with an additional UV treatment of effluents prior to release tions are present in large sample volumes the methods used in surface waters. for virus concentration can have significant influences. We Elimination of viruses in WWTPs depend on the char- therefore compared the detection rates obtained by ultra- acteristic features of the viruses as well as on the structures centrifugation and PEG precipitation, using samples with or and combinations of treatment steps of the plants. Further- without virus concentration steps (Fig. 3). Direct sampling more, there is a lack of data for HEV RNA reduction during and ultracentrifugation revealed comparable monthly detec- treatment in WWTPs. So far, reports with quantified HEV tion rates, whereas the PEG-processed samples resulted in concentrations in environmental waters are rare and mainly lower HEV RNA findings. Direct virus detection is easy to restricted to wastewater influent samples. The HEV con- perform but since small volumes are used, the limit of detec- centrations in wastewater influents determined in this study tion is much higher than for methods with virus concentra- correspond to the concentrations reported by Rodriguez- tion steps. Due to the small tested volumes, calculated virus Manzano et al. (2010), Masclaux et al. (2013), Wang et al. concentrations could be over- or underestimated. Therefore, (2018b) and Miura et al. (2016). In contrast to the findings when samples are tested by different methods, final virus of Masclaux et al. (2013), where HEV RNA was detected concentration should always be reported together with the more frequently in summer, no clear seasonal pattern of concentration method to ensure valid comparisons of the HEV RNA occurrence was observed in the present study, obtained data. similar to the report of Ram et al. (2016). Ultracentrifugation and PEG precipitation are standard Since wastewater effluents are discharged into rivers, virus concentration methods for detection of HEV in sewage further investigations were carried out in two urban rivers samples (Clemente-Casares et al. 2003; Rodriguez-Manzano (Table 2). Under normal weather conditions, about 30% of et al. 2010; Masclaux et al. 2013; Myrmel et al. 2015; Ram these river samples were positive with low concentrations of et al. 2016; Iaconelli et al. 2017; Wang et al. 2018b; Matos HEV RNA. However, the median HEV RNA concentration et al. 2018). The 25% positive samples during the 1-year of all tested river samples was below the LOD. These low period using the PEG precipitation method were comparable detection rates and low RNA concentrations are reasonable to findings of Masclaux et al. (2013) and Miura et al. (2016), due to virus dilution in big water volumes. Similar results which reported rates of 32% and 22%, respectively. have been reported in Italian surface waters impacted by In our hands, detection rates were much higher using the runoffs from grazing land and discharges from treatment ultracentrifugation method. This method does not need any plants, where 25% of the tested water samples were HEV addition of chemicals and since viruses tend to attach to RNA-positive (Idolo et al. 2013). In line with the results suspended matter (Jin and Flury 2002), ultracentrifugation from the two rivers under normal conditions, only 15% of is a suitable method for influent samples of WWTPs. Moreo- samples from a bathing water were positive for HEV RNA, ver, these ultracentrifugated samples were most suited for with a median concentration of all tested samples below the genotyping of HEV strains. LOD. Therefore, no evidence of an increased health risk In contrast to quantitative RT-PCR detection, longer was found at this bathing area. This reflects the water man- fragments need to be amplified for genotyping, which may agement efforts to maintain the bathing water quality under result in a lower sensitivity. In the present study, 15 ampli- normal weather conditions. cons (9%) of the WWTP influent samples could be success- In urban areas, mixed channels for sewage and rain water fully sequenced. Using nested RT-PCR, similar results of may reach capacity limits after heavy rainfall events and 5–13.5% of influent samples positive for HEV RNA were lead to release of uncleared wastewater into rivers (com- reported from Italy and Spain (Rusiñol et al. 2015; Iaconelli bined sewer overflows, CSOs). et al. 2017; Alfonsi et al. 2018). For genotyping, samples Such CSOs seem to have a high impact on HEV detec- with virus concentrations steps were most suitable, since 13 tion rates and concentrations in rivers, as seen in this study amplicons were sequenced from ultracentrifugated or PEG- for river 1, where several CSO sites are located. After three precipitated samples. Although the nested primer system heavy rainfall events in summer 2016 causing CSOs in river was able to amplify all HEV genotypes (Johne et al. 2010), 1, HEV positive samples increased from 33 to 75%, with a only HEV-3 strains were detected in environmental water median copy number of 2 × 10 copies/100 ml. Therefore, samples. The most prevalent HEV subtype was HEV-3c. In urban rivers may contain high HEV RNA concentrations addition, two samples contained HEV genotype 3f. HEV- during rainfall-affected periods, thereby increasing the pub- 3c and HEV-3f were also recently reported in wastewater lic health risk of HEV infections over the faecal-oral route in Italy (Di Profio et al. 2019). Of our three samples which by bathing or recreational activities in the polluted urban could not be subtyped exactly, two are most likely of subtype rivers. 3a and one of subtype 3c or 3i. 1 3 Food and Environmental Virology (2020) 12:137–147 145 carefully timed representative samples of combined sewer overflow The detected environmental HEV genotypes correlate events. well with reported subgenotype data from clinical samples from Germany (Vollmer et al. 2012; Tabatabai et al. 2014; Open Access This article is licensed under a Creative Commons Attri- Adlhoch et al. 2016). Genotype HEV-3c was reported to bution 4.0 International License, which permits use, sharing, adapta- be the most prevalent genotype in German blood donors, tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, and genotypes 3a and 3e were also found in clinical sam- provide a link to the Creative Commons licence, and indicate if changes ples (Vollmer et al. 2012). Moreover, HEV-3c was identi- were made. The images or other third party material in this article are fied in the first German clinical report of acute hepatitis E included in the article’s Creative Commons licence, unless indicated during pregnancy (Tabatabai et al. 2014) and is the most otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not common type in the European Union/ European Economic permitted by statutory regulation or exceeds the permitted use, you will Area (EFSA 2017). need to obtain permission directly from the copyright holder. To view a Most of the wastewater influent samples were obtained from copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. an urban WWTP with a catchment area of about 1.1 million people of the Berlin area. A recent study investigated HEV genotype 3 variants in patients from the same area and iden- References tified subtype 3c as the most prevalent HEV strain, besides genotypes 3e and 3f (Wang et al. 2018a). In our study, subtype Adlhoch, C., Avellon, A., Baylis, S. A., Ciccaglione, A. R., Couturier, E., de Sousa, R., et al. (2016). Hepatitis E virus: assessment of the HEV 3c was also detected in the suburban WWTP 3 with a epidemiological situation in humans in Europe, 2014/15. Journal pig farm located nearby. However, the overall HEV detection of Clinical Virology, 82, 9–16. rate did not differ from the other WWTPs. Besides domestic Alfonsi, V., Romanò, L., Ciccaglione, A. R., La Rosa, G., Bruni, R., pigs, in which HEV infection is highly prevalent (Fernández- Zanetti, A., et al. (2018). Hepatitis E in Italy: 5 years of national epidemiological, virological and environmental surveillance, 2012 Barredo et al. 2007; Jiménez de Oya et al. 2011; Dremsek et al. to 2016. Eurosurveillance, 23(41), 1700517. 2013), wild boars have been identified as a possible source Clemente-Casares, P., Pina, S., Buti, M., Jardi, R., Martín, M., Bofill- of HEV RNA. In addition, it has to be considered that Ber- Mas, S., et al. (2003). Hepatitis E Virus epidemiology in indus- lin also harbours a large population of wild boars (Stillfried trialized Countries. Emerging Infectious Diseases, 9(4), 449–454. Clemente-Casares, P., Ramos-Romero, C., Ramirez-Gonzalez, E., et al. 2017). In a current study, a high degree of nucleotide & Mas, A. (2016). Hepatitis E virus in industrialized countries: sequence homology in a wild boar isolate and a human isolate the silent threat. BioMed Research International. https ://doi. was detected and zoonotic HEV-3c and 3a were identified in org/10.1155/2016/98380 41. wild boars in the Berlin/Potsdam area (Schielke et al. 2009), Dalton, H. R., Kamar, N., & Izopet, J. (2014). Hepatitis E in developed countries: Current status and future perspectives. Future Microbi- corresponding with the detection of HEV-3c and 3a RNA in ology, 9(12), 1361–1372. this study. In cities, HEV transmission from wild boars to Di Profio, F., Melegari, I., Palombieri, A., Sarchese, V., Arbuatti, A., humans has to be taken into consideration, either by direct or Fruci, P., et al. (2019). High prevalence of hepatitis E virus in raw by indirect transmission by surface waters, environment, or sewage in Southern Italy. Virus Research, 272, 197710. Dremsek, P., Joel, S., Baechlein, C., Pavio, N., Schielke, A., Ziller, M., other carrier animals (Schielke et al. 2009). et al. (2013). Hepatitis E virus seroprevalence of domestic pigs The composition of subtypes detected in the environmen- in Germany determined by a novel in-house and two reference tal waters in the present study reflects the circulation of HEV ELISAs. Journal of Virological Methods, 190(1–2), 11–16. strains in humans and animals in the same region. Previous EFSA Panel on Biological Hazards (BIOHAZ), Ricci, A., Allende, A., Bolton, D., Chemaly, M., Davies, R., et al. (2017). Public health reports from other countries also demonstrated a correlation risks associated with hepatitis E virus (HEV) as a food-borne between detected viruses in wastewater and clinical cases pathogen. EFSA Journal, 15(7), e04886. (Hellmér et al. 2014; Ivanova et al. 2019). Faber, M., Willrich, N., Schemmerer, M., Rauh, C., Kuhnert, R., Stark, Detection of viruses in environmental waters can con- K., et al. (2018a). Hepatitis E virus seroprevalence, seroincidence and seroreversion in the German adult population. Journal of tribute to a better understanding of the epidemiology and Viral Hepatitis, 25(6), 752–758. prevalent strains in the population. Although foodborne Faber, M., Askar, M., & Stark, K. (2018b). Case-control study on risk transmission is considered as the main pathway of zoonotic factors for acute hepatitis E in Germany, 2012 to 2014. Eurosur- HEV infection, environmental transmission should not be veillance. https ://doi.org/10.2807/1560-7917.ES.2018.23.19.17- 00469 . neglected and water monitoring should be integrated in the Fenaux, H., Chassaing, M., Berger, S., Gantzer, C., Bertrand, I., & One Health approach to reduce public health risks. Schvoerer, E. (2019). Transmission of hepatitis E virus by water: An issue still pending in industrialized countries. Water Research, Acknowledgements Open Access funding provided by Projekt DEAL. 151, 144–157. S.B. was supported by a doctoral fellowship by the UBA. We thank F. Fernández-Barredo, S., Galiana, C., García, A., Gómez-Muñoz, M. T., Kubek (UBA) and the WWTP operating companies for wastewater Vega, S., Rodríguez-Iglesias, M. A., et al. (2007). Prevalence and samples. In addition, we are grateful to our partners in the BMBF genetic characterization of hepatitis E virus in paired samples of financed project Flusshygiene (02WRM1364B), especially W. Seis from the Berlin Centre of Competence for Water (KWB) for collecting 1 3 146 Food and Environmental Virology (2020) 12:137–147 feces and serum from naturally infected pigs. Canadian Journal Manor, Y., Blomqvist, S., Sofer, D., Alfandari, J., Halmut, T., Abra- of Veterinary Research, 71(3), 236. movitz, B., et al. (2007). Advanced environmental surveillance Garson, J. A., Ferns, R. B., Grant, P. R., Ijaz, S., Nastouli, E., Szypul- and molecular analyses indicate separate importations rather ska, R., et al. (2012). Minor groove binder modification of widely than endemic circulation of wild type 1 poliovirus in Gaza dis- used TaqMan probe for hepatitis E virus reduces risk of false trict in 2002. Applied and Environment Microbiology, 73(18), negative real-time PCR results. Journal of Virological Methods, 5954–5958. 186(1–2), 157–160. Masclaux, F. G., Hotz, P., Friedli, D., Savova-Bianchi, D., & Hellmér, M., Paxéus, N., Magnius, L., Enache, L., Arnholm, B., Oppliger, A. (2013). High occurrence of hepatitis E virus in Johansson, A., et al. (2014). Detection of pathogenic viruses in samples from wastewater treatment plants in Switzerland and sewage provided early warnings of hepatitis A virus and norovi- comparison with other enteric viruses. Water Research, 47(14), rus outbreaks. Applied and Environmental Microbiology, 80(21), 5101–5109. 6771–6781. Matos, A., Mesquita, J. R., Gonçalves, D., Abreu-Silva, J., Luxo, C., Iaconelli, M., Muscillo, M., Della Libera, S., Fratini, M., Meucci, L., & Nascimento, M. S. J. (2018). First detection and molecular De Ceglia, M., et al. (2017). One-year surveillance of human characterization of hepatitis E virus in water from wastewater enteric viruses in raw and treated wastewaters, downstream river treatment plants in Portugal. Annals of Agricultural and Envi- waters, and drinking waters. Food and Environmental Virology, ronmental Medicine, 25(2), 364–367. 9(1), 79–88. Meng, X. J. (2010). Hepatitis E virus: animal reservoirs and zoonotic Idolo, A., Serio, F., Lugoli, F., Grassi, T., Bagordo, F., Guido, M., risk. Veterinary Microbiology, 140(3–4), 256–265. et al. (2013). Identification of HEV in symptom-free migrants Miura, T., Lhomme, S., Le Saux, J. C., Le Mehaute, P., Guillois, and environmental samples in Italy. The Journal of Viral Hepa- Y., Couturier, E., et al. (2016). Detection of hepatitis E virus titis, 20, 438–443. in sewage after an outbreak on a French island. Food and Envi- Ivanova, O. E., Yarmolskaya, M. S., Eremeeva, T. P., Babkina, G. ronmental Virology, 8(3), 194–199. M., Baykova, O. Y., Akhmadishina, L. V., et al. (2019). Envi- Myrmel, M., Lange, H., & Rimstad, E. (2015). A 1-year quantitative ronmental surveillance for poliovirus and other enteroviruses: survey of noro-, adeno-, human boca-, and hepatitis E viruses in Long-term experience in Moscow, Russian Federation, 2004– raw and secondarily treated sewage from two plants in Norway. 2017. Viruses, 11(5), 424. Food and Environmental Virology, 7(3), 213–223. Jiménez de Oya, N., de Blas, I., Blázquez, A. B., Martín-Acebes, M. Pavio, N., Doceul, V., Bagdassarian, E., & Johne, R. (2017). Recent A., Halaihel, N., Gironés, O., et al. (2011). Widespread distribu- knowledge on hepatitis E virus in Suidae reservoirs and trans- tion of hepatitis E virus in Spanish pig herds. BMC Research mission routes to human. Veterinary Research, 48(1), 78. Notes, 4(1), 412. Pérez-Gracia, M. T., García, M., Suay, B., & Mateos-Lindemann, M. Jin, Y., & Flury, M. (2002). Fate and transport of viruses in porous L. (2015). Current knowledge on hepatitis E. Journal of Clinical media. In D. Sparks (Ed.), Advances in agronomy (Vol. 77, pp. and Translational Hepatology, 3(2), 117. 39–102). San Diego: Academic Press. Pina, S., Jofre, J., Emerson, S. U., Purcell, R. H., & Girones, R. Johne, R., Plenge-Boenig, A., Hess, M., Ulrich, R. G., Reetz, J., & (1998). Characterization of a strain of infectious hepatitis E Schielke, A. (2010). Detection of a novel hepatitis E-like virus virus isolated from sewage in an area where hepatitis E is not in faeces of wild rats using a nested broad-spectrum RT-PCR. endemic. Applied and Environment Microbiology, 64(11), Journal of General Virology, 91(3), 750–758. 4485–4488. Johne, R., Reetz, J., Ulrich, R. G., Machnowska, P., Sachsenröder, J., Ram, D., Manor, Y., Gozlan, Y., Schwartz, E., Ben-Ari, Z., Mendel- Nickel, P., et al. (2014). An ORF 1-rearranged hepatitis E virus son, E., et al. (2016). Hepatitis E virus genotype 3 in sewage derived from a chronically infected patient efficiently replicates and genotype 1 in acute hepatitis cases, Israel. The American in cell culture. Journal of Viral Hepatitis, 21(6), 447–456. Journal of Tropical Medicine and Hygiene, 95(1), 216–220. Jothikumar, N., Cromeans, T. L., Robertson, B. H., Meng, X. J., & RKI. (2019). Infektionsepidemiologisches Jahrbuch meldepflichtiger Hill, V. R. (2006). A broadly reactive one-step real-time RT- Krankheiten für 2018. Berlin, Germany: Robert Koch-Institut. PCR assay for rapid and sensitive detection of hepatitis E virus. Rodriguez-Manzano, J., Miagostovich, M., Hundesa, A., Clemente- Journal of Virological Methods, 131(1), 65–71. Casares, P., Carratala, A., Buti, M., et al. (2010). Analysis of Kimura, M. (1980). A simple method for estimating evolution- the evolution in the circulation of HAV and HEV in eastern ary rates of base substitutions through comparative studies of Spain by testing urban sewage samples. Journal of Water and nucleotide sequences. Journal of Molecular Evolution, 16(2), Health, 8(2), 346–354. 111–120. Rusiñol, M., Fernandez-Cassi, X., Timoneda, N., Carratalà, A., Krumbholz, A., Neubert, A., Joel, S., Girschick, H., Huppertz, H. I., Abril, J. F., Silvera, C., et  al. (2015). Evidence of viral dis- Kaiser, P., et al. (2014). Prevalence of hepatitis E virus antibod- semination and seasonality in a Mediterranean river catchment: ies in children in Germany. The Pediatric Infectious Disease implications for water pollution management. Journal of Envi- Journal, 33(3), 258–262. ronmental Management, 159, 58–67. Kumar, S., Stecher, G., & Tamura, K. (2016). MEGA7: molecular Rutjes, S. A., Lodder, W. J., Lodder-Verschoor, F., Van den Berg, H. evolutionary genetics analysis version 70 for bigger datasets. H., Vennema, H., Duizer, E., et al. (2009). Sources of hepatitis Molecular Biology and Evolution, 33(7), 1870–1874. E virus genotype 3 in The Netherlands. Emerging Infectious La Rosa, G., Fontana, S., Di Grazia, A., Iaconelli, M., Pourshaban, Diseases, 15(3), 381. M., & Muscillo, M. (2007). Molecular identification and genetic Schielke, A., Sachs, K., Lierz, M., Appel, B., Jansen, A., & Johne, R. analysis of norovirus genogroups I and II in water environ- (2009). Detection of hepatitis E virus in wild boars of rural and ments: comparative analysis of different reverse transcription- urban regions in Germany and whole genome characterization PCR assays. Applied and Environment Microbiology, 73(13), of an endemic strain. Virology Journal, 6(1), 58. 4152–4161. Smith, D. B., Paddy, J. O., & Simmonds, P. (2016a). The use of La Rosa, G., Pourshaban, M., Iaconelli, M., Vennarucci, V. S., & human sewage screening for community surveillance of hepa- Muscillo, M. (2010). Molecular detection of hepatitis E virus titis E virus in the UK. Journal of Medical Virology, 88(5), 915–918. in sewage samples. Applied and Environment Microbiology, 76(17), 5870–5873. 1 3 Food and Environmental Virology (2020) 12:137–147 147 Smith, D. B., Simmonds, P., Izopet, J., Oliveira-Filho, E. F., Ulrich, E virus infection in German blood donors. Journal of Clinical R., Johne, R., et al. (2016b). Proposed reference sequences for Microbiology, 50(8), 2708–2713. hepatitis E virus subtypes. The Journal of General Virology, Wang, B., Harms, D., Papp, C. P., Niendorf, S., Jacobsen, S., Lütge- 97(3), 537. hetmann, M., et al. (2018a). Comprehensive molecular approach Sridhar, S., Lau, S. K., & Woo, P. C. (2015). Hepatitis E: A disease for characterization of hepatitis E virus genotype 3 variants. of reemerging importance. Journal of the Formosan Medical Journal of Clinical Microbiology, 56(5), e01686–e1717. Association, 114(8), 681–690. Wang, H., Sikora, P., Rutgersson, C., Lindh, M., Brodin, T., Björle- Stillfried, M., Fickel, J., Börner, K., Wittstatt, U., Heddergott, M., nius, B., et al. (2018b). Differential removal of human patho- Ortmann, S., et al. (2017). Do cities represent sources, sinks genic viruses from sewage by conventional and ozone treat- or isolated islands for urban wild boar population structure? ments. International Journal of Hygiene and Environmental Journal of Applied Ecology, 54(1), 272–281. Health, 221(3), 479–488. Tabatabai, J., Wenzel, J. J., Soboletzki, M., Flux, C., Navid, M. H., & Schnitzler, P. (2014). First case report of an acute hepatitis E Publisher’s Note Springer Nature remains neutral with regard to subgenotype 3c infection during pregnancy in Germany. Journal jurisdictional claims in published maps and institutional affiliations. of Clinical Virology, 61(1), 170–172. Vollmer, T., Diekmann, J., Johne, R., Eberhardt, M., Knabbe, C., & Dreier, J. (2012). Novel approach for detection of hepatitis 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Food and Environmental Virology Springer Journals

Detection and Characterization of Hepatitis E Virus Genotype 3 in Wastewater and Urban Surface Waters in Germany

Loading next page...
 
/lp/springer-journals/detection-and-characterization-of-hepatitis-e-virus-genotype-3-in-0lC0Hnr0XR
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2020
ISSN
1867-0334
eISSN
1867-0342
DOI
10.1007/s12560-020-09424-2
Publisher site
See Article on Publisher Site

Abstract

In highly populated areas, environmental surveillance of wastewater and surface waters is a key factor to control the circu- lation of viruses and risks for public health. Hepatitis E virus (HEV) genotype 3 is considered as an emerging pathogen in industrialized countries. Therefore, this study was carried out to determine the prevalence of HEV in environmental waters in urban and suburban regions in Germany. HEV was monitored in water samples using quantitative RT-PCR (RT-qPCR) and nested RT-PCR without or with virus concentration via polyethylene glycol precipitation or ultracentrifugation. By RT- qPCR, 84–100% of influent samples of wastewater treatment plants were positive for HEV RNA. Genotypes HEV-3c and 3f were identified in wastewater, with HEV-3c being the most prevalent genotype. These data correlate with subtypes identified earlier in patients from the same area. Comparison of wastewater influent and effluent samples revealed a reduction of HEV RNA of about 1 log during passage through wastewater treatment plants. In addition, combined sewer overflows (CSOs) after heavy rainfalls were shown to release HEV RNA into surface waters. About 75% of urban river samples taken during these CSO events were positive for HEV RNA by RT-qPCR. In contrast, under normal weather conditions, only around 30% of river samples and 15% of samples from a bathing water located at an urban river were positive for HEV. Median concentrations of HEV RNA of all tested samples at this bathing water were below the limit of detection. Keywords Hepatitis E virus · Monitoring · Genotyping · Wastewater · Surface water · Combined sewer overflow Introduction Genotypes 3 and 4 are zoonotic and infect mainly humans, swine and wild boars (Pavio et al. 2017). Whereas genotype Hepatitis E virus (HEV) is the causative agent of acute 4 is mainly restricted to Asia, in most industrialized coun- and chronic hepatitis in humans worldwide. A severe dis- tries genotype 3 is predominant (Clemente-Casares et al. ease progression is possible with mortality rates around 2003; Meng 2010; Dalton et al. 2014). 1% (Pérez-Gracia et al. 2015). However, among pregnant In Germany seroprevalences for HEV-specific antibodies women infected with HEV genotype 1, a higher incidence of about 1% in children (Krumbholz et al. 2014) and about and severity was observed with mortality rates up to 30% 15% among adults (Faber et al. 2018a) were reported. A (Clemente-Casares et al. 2016). continuous increase in the number of notified hepatitis E HEV is classified into four main human-pathogenic gen- cases was recorded in Germany during the last years, most otypes within the Hepeviridae family. Genotypes 1 and 2 likely due to increased awareness. In 2018 about 3400 new infect only humans and are endemic in developing countries. hepatitis E cases were reported to the Robert Koch Institute (RKI 2019). HEV is transmitted mainly through meat products of * Hans-Christoph Selinka infected animals and faecally contaminated water. Trans- hans-christoph.selinka@uba.de mission to humans through contaminated water is known for Section II 1.4 Microbiological Risks, German Environment genotypes HEV-1 and HEV-2, mainly in developing coun- Agency (UBA), Corrensplatz 1, 14195 Berlin, Germany tries (Fenaux et al. 2019). Berliner Wasserbetriebe (BWB), Cicerostr. 24, 10709 Berlin, Industrialization of a country decreases HEV risk related Germany to HEV-1, but increases that related to HEV-3 and HEV-4, German Federal Institute for Risk Assessment (BfR), as observed in China (Sridhar et al. 2015). Max-Dohrn-Straße 8-10, 10589 Berlin, Germany Vol.:(0123456789) 1 3 138 Food and Environmental Virology (2020) 12:137–147 So far, the role of water in the transmission of zoonotic taken by the Berlin Centre of Competence for Water during HEV-3 has only been suspected (Fenaux et al. 2019). A a sampling campaign after heavy rainfall events in 2016. recent study identified the occupational contact with waste- Additional samples were drawn and analysed from a bathing water as a risk factor associated with autochthonous hepatitis water located at river 2 (river 2/bathing water) in the years E in Germany, supporting that waterborne transmission of 2018 and 2019. Water samples were processed directly after HEV-3 is possible (Faber et al. 2018b). sampling or stored at − 80 °C until further processing. In developed countries human and animal hosts of HEV-3 may contaminate wastewater through their faeces. HEV par- Sample Concentration ticles can reach the environment and potentially contaminate surface waters. Thus, surface waters could be a source of In environmental samples, human-pathogenic viruses are HEV contamination for animals and humans (Fenaux et al. mostly present in low or very low concentrations and have 2019). to be further concentrated for analyses. In our study we used Increasing HEV prevalence in industrialized countries is ultracentrifugation (U) and polyethylene glycol (PEG) pre- known since 1998. In Spain, HEV detection in urban sewage cipitation (Fig. 1). samples was reported (Pina et al. 1998), followed by reports Ultracentrifugation was performed according to a pre- from the Netherlands (Rutjes et al. 2009), Italy (La Rosa viously described method (La Rosa et al. 2007). In brief, et al. 2010) and other countries. Most common detection 180 ml supernatant after initial centrifugation at 3000×g for methods are nested reverse transcription (RT) polymerase 10 min was pelleted by 2 h centrifugation at 160,000×g with chain reaction (PCR) or quantitative RT-PCR (RT-qPCR) a 45Ti rotor in an Optima L-100 K ultracentrifuge (Beck- with or without prior virus concentration steps. In recent mann, Germany) and resuspended in 5 ml PBS for nucleic years, Italy, Norway and the UK have reported first investi- acid extraction. gations for a HEV surveillance in sewage (Idolo et al. 2013; For virus concentration by PEG precipitation (Manor Myrmel et al. 2015; Smith et al. 2016a; Alfonsi et al. 2018). et al. 2007), PEG 6000 (80 g) and NaCl (17.5 g) were added To the best of our knowledge there are no available stud- to 1 l water samples, mixed for 1 h and stored overnight ies on the presence of HEV in environmental waters in Ger- at 4 °C. Subsequently, precipitates containing the viruses many. Therefore, this study was carried out to (1) investigate were collected after 1 h centrifugation at 12,200×g. Pel- the HEV prevalence in environmental water samples, (2) lets were resuspended in 15 ml PBS, 15 ml chloroform (to to compare HEV concentration methods, and (3) to geno- destroy bacteria) and 150 µl Tween-80. After centrifuga- type detected HEV strains. Wastewater influent and efflu- tion for 15 min at 1400×g, the top layer was saved and the ent samples of urban and suburban wastewater treatment lower chloroform layer was removed. Remaining pellets plants (WWTPs), surface waters from two rivers including were resuspended in 0.05 mol/l glycine pH 7.2 with 3% beef a bathing water and conditions of combined sewer overflows extract. Centrifugation was repeated and both supernatants (CSO) were investigated. For comparison of virus concen- tration techniques, PEG-precipitated samples, samples sub- jected to ultracentrifugation and samples without further virus concentrations were tested simultaneously. Genotyping was performed for further characterization of the detected HEV strains. Material and Methods Sampling Samples of wastewater influents (after coarse grid removal) and wastewater effluents (secondary effluents, before UV treatment) of WWTPs were collected in the years 2014–2019 from central urban (WWTP 1) and suburban (WWTPs 2–4) areas of the cities of Berlin and Munich, Germany. Surface water samples were taken in the years 2016–2019 from two urban rivers at normal weather conditions (river 1 and river 2), as well as after heavy rainfall events with com- Fig. 1 Flow chart of methods applied on wastewater and river water bined sewer overflows (river 1/CSO). CSO samples were samples for HEV RNA detection 1 3 Food and Environmental Virology (2020) 12:137–147 139 were combined to a final volume of about 20 ml. Exact vol- not detected, the LOD concentration was used for further umes were noted for calculating virus concentrations in the calculations. With 200 µl of viral nucleic acids eluted by the original samples. NucliSENS® easyMAG® method from 5 ml of direct water samples, the LOD was 200 copies/100 ml. Using nucleic Nucleic Acid Extraction acids from water samples concentrated by ultracentrifuga- tion, an LOD of 6 copies/100 ml was achieved and the LOD Nucleic acid extraction was performed with 5 ml volumes of PEG-precipitated samples was four copies per 100 ml. of concentrated samplesor 5 ml volumes of samples with- The limit of quantification (LOQ) was set ten times higher out further virus concentration steps (direct samples). The than the LOD of each method. NucliSENS® easyMAG® (bioMérieux, Germany) method allows simultaneous extraction of DNA and RNA with HEV‑Specific Nested RT‑PCR same efficiencies. To assess the extraction efficacy, a sam- ple spiked with human adenovirus 2 with a defined con- Primers for nested RT-PCR, which amplify a 332 bp product centration was included for each set of samples subjected from the HEV open reading frame 1 (ORF1) were designed to the nucleic acid extraction procedure. The method was by Johne et al. (2010). RNA from the HEV isolate 47832c used according to the manufacturer’s protocol, with slight (Johne et al. 2014) was used as positive control. For the modifications. Centrifugation was performed at 6000×g for first RT-PCR with a total reaction volume of 25 µl, 5 µl of 5 min after lysis buffer incubation to eliminate large disturb- template was amplified using the OneStep Ahead RT-PCR ing particles present in turbid water samples. In addition, Kit (QIAGEN, Germany). Cycling profile included the fol- purified nucleic acids were eluted two times in 100 µl elution lowing settings: 10 min at 50 °C, 5 min at 95 °C, 40 cycles buffer resulting in a final volume of 200 µl to allow analyses of 10 s at 95 °C, 10 s at 55 °C, 10 s at 72 °C and 2 min at of several qPCR reactions. 72 °C. The second nested PCR was performed with 2 µl template from the first RT-PCR. The Taq DNA Polymerase HEV‑Specific Quantitative Real‑Time RT‑PCR Kit (QIAGEN, Germany) was used according to protocol in a total reaction volume of 50 µl. Primer concentrations were Primers designed by Jothikumar et al. (2006) were used for 0.3 µM and cycling conditions were the following: 3 min quantitative RT-PCR. Probes were used either as described at 94 °C, 35 cycles of 45 s at 94 °C, 45 s at 60 °C, 1 min at by Jothikumar et al. (2006) or in a modified version (Gar - 72 °C and 10 min at 72 °C. son et al. 2012) using the quencher MGB (Minor groove PCR fragments were separated by gel electrophoresis on binder) to reduce the risk of false negative real-time RT-PCR 1.5% agarose gels in 1 × TBE buffer with 10 µl of 10,000×g results. RT-qPCR was performed in a volume of 25 µl using GelRed staining (Biotium, Germany) per 100 ml agarose the QuantiTect Probe RT-PCR Kit (QIAGEN, Germany) solution. Loading buffer (Thermo Scientific, Germany) was with a probe concentration of 0.2 µM. The following cycle mixed with the PCR products and gels were run for 50 min conditions were applied: 30 min at 50 °C, 15 min at 95 °C at 90 V. Low range DNA ladder (5 µl) was used as a size and 45 cycles with 15 s at 94 °C and 1 min at 56 °C. Each marker (Thermo Scientific, Germany). reaction mix contained 10 µl of undiluted or 1:10 diluted templates (4 reactions per sample) to detect putative inhibi- DNA Sequencing and Nucleotide Sequence Analyses tion of the RT-qPCR reaction. Copy numbers are calculated based on all reactions, if undiluted and diluted samples cor- Bands of the expected length (332 bp) were excised and respond. In the case of partial inhibition in the undiluted purified according to the protocol from innuPREP DOU- samples, copy numbers of the diluted samples were chosen BLEpure Kit (Analytik Jena, Germany). The cDNA was for quantification. eluted twice in 30 µl elution buffer, combined and sequenced Double-stranded DNA Gene Strands (Eurofins Genom- by Eurofins (Germany). ics, Sweden) containing the specific amplification sequence All HEV sequences determined in this study were were applied as quantitative HEV standards in concentra- submitted to NCBI GenBank under accession numbers 6 1 tions from 10 HEV copies/10 µl to 10 HEV copies/10 µl MT087290 to MT087304. Sequence alignments and to generate a standard curve for determination of virus copy phylogenetic trees were constructed with Molecular numbers in the samples. Standard deviations of samples Evolutionary Genetics Analysis Version 7.0 (MEGA 7) during the 1-year surveillance were calculated from two to software (Kumar et  al. 2016). The MUSCLE program seven monthly samples. The calculation of the limit of detec- was used for multiple sequence alignment and maxi- tion (LOD) of HEV RNA was based on duplicates of 10 µl mum likelihood as statistic method based on the Kimura nucleic acid templates per RT-qPCR reaction with at least 1 2-parameter model (Kimura, 1980). The phylogenetic HEV copy to be detected in the duplicate. If HEV RNA was trees were validated by replicating with 1000 bootstraps. 1 3 140 Food and Environmental Virology (2020) 12:137–147 Obtained HEV sequences were aligned to 41 HEV-sub- Results type reference sequences (or a subset of 19 genotype 3 reference subtype sequences), as recommended by Smith Monitoring of HEV in Environmental Water Samples et al. (2016b). In addition, sequences were aligned to the by RT‑qPCR HEV-3c positive control (isolate 47832c from Johne et al. 2014) and 17 sequences from HEV infected patients from Surface waters as well as influent and effluent wastewater the Charité Hospital in Berlin (Wang et al. 2018a). samples of WWTPs from urban and suburban areas were monitored for HEV by RT-qPCR. Four urban and suburban WWTPs, differing in their catchment areas and cleaning Statistical Analyses capacities, were investigated (Table  1). Of 111 wastewa- ter influent samples collected in the urban WWTP 1, 84% Statistical analyses were performed with Microsoft Excel. were positive for HEV RNA with a median concentration of As quantification data are not normally distributed but 3 × 10 copies/100 ml. The median concentration of all 111 ordinally scaled, the Mann–Whitney U test was used to tested samples was 2 × 10 copies/100 ml. determine the statistical significance at a 95% confidence HEV RNA was also detected in 26 out of 83 wastewater level. This test was carried out to evaluate statistical dif- effluent samples (31%) of WWTP 1 with a median concen- ferences between monthly virus concentrations of WWTP tration of 1 × 10 copies/100 ml in positive samples. How- inf luent samples and WWTP eff luent samples (Fig. 2) as ever, the median concentration of all 83 tested effluent sam- well as between different virus concentrations methods ples was below the limit of detection (LOD). (Fig. 3). In three suburban WWTPs (WWTP 2–4), HEV RNA was detected in 86–100% of the influent samples with median concentrations of positive samples in the range of 2 × 10 copies/100  ml–1 × 10 copies/100 ml. Effluent samples of Fig. 2 Comparison of HEV RNA concentrations in monthly influent samples (I) and efflu- ent samples (E) of WWTP 1, analysed by RT-qPCR without virus concentration steps. Black and grey bars represent meas- ured HEV concentrations above the LOD (open bars). LOD is the limit of detection with 200 copies/100 ml Fig. 3 Concentration of HEV RNA in monthly influent sam- ples of an urban WWTP over a period of one year. For com- parison of sampling methods direct samples (D) and samples concentrated by ultracentrifuga- tion (U) and PEG precipitation (P) were analysed by RT-qPCR. Black bars represent measured HEV concentrations above the LOD (open bars), which differ in each method 1 3 Food and Environmental Virology (2020) 12:137–147 141 Table 1 Detection of HEV RNA in WWTP influent and effluent samples by RT-qPCR Cleaning WWTP Influent samples WWTP Effluent samples capacity 3 Tested (n) Positive Median positive Median all Tested (n) Positive Median positive Median all [m /day] [n/ (%)] samples* samples* [n/ (%)] samples* samples* 3 3 3 WWTP 1 257.000 111 93 (84%) 3 × 10 2 × 10 83 26 (31%) 1 × 10 < LOD 3 3 2 WWTP 2 5.500 10 9 (90%) 2 × 10 1 × 10 2 1 (50%) 8 × 10 4 × 102 3 3 2 WWTP 3 119.000 7 6 (86%) 4 × 10 3 × 10 3 3 (100%) 4 × 10 4 × 102 4 4 WWTP 4 40.000 6 6 (100%) 1 × 10 1 × 10 nt nt nt nt WWTP wastewater treatment plant, LOD limit of detection, nt not tested *[HEV copies/100 ml] was 3 × 10 copies/100 ml, but the median concentration of Table 2 Detection of HEV RNA in surface water samples by RT- qPCR all 55 tested samples was below the LOD. Surface water samples One‑year HEV Surveillance of a Wastewater Tested Positive Median positive Median all Treatment Plant (n) [n/(%)] samples* samples* River 1 21 7 (33%) 6 × 10 < LOD To investigate if the high variability in the concentrations of 3 3 River 1/CSO 16 12 (75%) 2 × 10 2 × 10 HEV in influent and effluent samples was affected by envi- River 2 69 21 (30%) 9 × 102 < LOD ronmental or seasonal influences, the central urban WWTP River 2/BW 55 8 (15%) 3 × 102 < LOD 1 was surveilled during a complete cycle of a year (Fig. 2). In WWTPs, virus concentrations are sufficiently high LOD limit of detection, BW bathing water, CSO combined sewer to be detected in small volumes without further virus con- overflow centration steps. Therefore, direct samples were measured * [HEV copies/100 ml] from March 2018 to February 2019 in influent and effluent samples of WWTP 1. HEV RNA was detected in 11 out of 12 monthly influent samples and in eight out of 12 effluent suburban WWTPs were positive for HEV RNA at rates of samples based on the LOD of 200 copies/100 ml. 50% (WWTP 2) and 100% (WWTP 3). Median concentra- Influent and effluent samples were taken the same day tions of these positive samples were 8 × 10 copies/100 ml without considering the passage time of wastewater treat- and 4 × 10 copies/100 ml, respectively. Median concen- ment. Several samples were collected each month. The mean trations of all tested effluent samples of these suburban concentration of all monthly measured samples is shown in WWTPs were 4 × 10 copies/100 ml. the figure for each month and was used to compare influent The results for HEV monitoring of surface waters are and effluent samples of WWTP 1. shown in Table 2. About 30% of 90 tested samples of two The mean value of calculated HEV RNA concentrations urban rivers under normal weather conditions (river 1 and of 12 monthly influent samples over the surveilled year was river 2) were positive for HEV RNA with median concen- 2 × 10 genome copies/100 ml. Effluent samples resulted in 2 2 3 trations of 6 × 10 copies/100 ml and 9 × 10 copies/100 ml, a mean of 2 × 10 copies/100 ml over this one-year period. respectively. Although effluents of WWTP 1 are released The average HEV RNA reduction during the passage of the into river 2 about 3 miles upstream of the sampling site, WWTP was about 1 log , comparing influent and effluent median concentrations of all river samples were below the samples above the LOD. Moreover, HEV concentrations of LOD. However, after heavy rainfall events, causing com- influent samples are significantly higher than from effluent bined sewer overflows (CSOs) upstream into river 1 (river 1/ samples (Mann–Whitney U test, p < 0.05). With a limit of CSO), 75% of the samples were positive for HEV RNA with quantification (LOQ) set to tenfold LOD, 10 of 12 influent a median concentration of 2 × 10 copies/100 ml. samples and only 2 effluent samples were positive for HEV In a bathing water located at the urban river 2 (river 2/ RNA, demonstrating the clearing effect of at least 1 log in BW) downstream the first sampling site of river 2, only eight the wastewater treatment plant. During the surveilled year, out of 55 samples (15%) were positive for HEV. The median no obvious seasonal pattern of HEV occurrence in wastewa- concentration of eight positive samples of this bathing water ter samples was observed. 1 3 142 Food and Environmental Virology (2020) 12:137–147 1 3 Food and Environmental Virology (2020) 12:137–147 143 ◂Fig. 4 Characterization of HEV strains from wastewater samples by 94 wastewater effluent samples and 57 river water samples gel electrophoresis and sequencing. a Exemplary agarose gel with showed a clear 332 bp band in the nested RT-PCR suitable HEV positive samples (332 bp fragments) from two WWTP influent for sequencing (data not shown). samples. b Maximum likelihood phylogenetic consensus tree of HEV Genotyping and subtyping were performed by sequence strains detected in urban wastewaters. Numbers at the nodes repre- sent bootstrap values > 60. Scale bar indicates the genetic distance alignments with reference strains followed by phylogenetic (nucleotide substitutions per site). Identified HEV sequences detected analyses (Fig.  4b). All identified sequences belonged to in wastewater samples are marked with a black dot. Names consist HEV genotype 3. Therefore, only reference sequences of of accession numbers, places, months, years of sampling and prepa- genotype 3 are shown. Bootstrap values > 60 are reported. ration methods (D: direct sample, U: ultracentrifugation, PEG: poly- ethylene glycol precipitation). Sequences from HEV infected patients HEV genotype 3c was the most prevalent subtype detected are labelled with open dots. HEV sequences were aligned to 41 in10 wastewater influent samples. Two wastewater samples HEV-subtype reference sequences denoted by accession number, sub- were identified as HEV genotype 3f. For three other samples genotype and source of first detection. Since all identified sequences no subtypes were classified. Of these 15 genotyped HEV belonged to genotype HEV-3, only sequences of this genotype are shown. Three rabbit HEV-3 sequences were used as outgroup strains from wastewater samples, ten were obtained from samples concentrated by ultracentrifugation, three from sam- ples prepared by PEG precipitation and two from samples Comparison of HEV Concentration Methods without further virus concentration steps. Moreover, the HEV genotypes identified in samples of urban and subur - To investigate if sample preparation methods have an impact ban WWTPs from the years 2016–2019 were compared to sequences of HEV infected patients from the same area from on the detection rate and the measured HEV RNA concen- trations, direct samples and samples concentrated by ultra- 2009–2016 (Wang et al. 2018a). As seen in the phylogenetic tree (Fig. 4), most of the wastewater and patient sequences centrifugation and polyethylene glycol precipitation were compared over a cycle of one year using influent samples of cluster in subtype 3c or 3f. WWTP 1 (Fig. 3). Each of these three methods has a different limit of Discussion detection, namely 4 copies/100 ml or 6 copies/100 ml for PEG precipitation and ultracentrifugation, respectively, or This study presents a quantitative surveillance and geno- 200 copies/100 ml for direct samples. In direct wastewater influent samples, HEV RNA was detected in 11 out of 12 typing of HEV strains in urban and suburban wastewater influent and effluent samples as well as in surface waters. monthly samples. In samples concentrated by ultracentrifu- gation, HEV RNA was found each month. Calculated HEV The zoonotic genotype 3 of HEV is autochthonous in many industrialized countries (Clemente-Casares et  al. concentrations in direct samples and samples concentrated by ultracentrifugation were in the similar range, in contrast 2003; Meng 2010; Dalton et al. 2014). Besides foodborne transmission of this genotype, environmental transmission to PEG-processed samples, which resulted in lower HEV RNA concentrations and lower detection rates. Using the pathways have also been proposed. In the present study, a wide distribution of HEV RNA in environmental waters in PEG method, viruses were detected only three times during this surveillance year and thus they were clearly signic fi antly Germany was identified, which may pose a risk of environ- mental transmission of HEV. However, HEV RNA detected different from direct samples and samples concentrated by ultracentrifugation (Mann–Whitney U test, p < 0.05). by PCR methods does not necessarily represent intact and infective virus particles. Genotyping of HEV from Environmental Water The highest detection rates (84–100%) of HEV RNA by quantitative PCR were found in wastewater influent sam- Samples ples, with a detection rate of 84% in WWTP 1 (Table 1). In contrast, only 31% of the effluent samples of WWTP 1 were To characterize the detected HEV strains in more detail, sequencings were carried out to identify HEV genotypes and positive for HEV RNA, demonstrating a cleaning effect of the WWTP with regard to HEV. In accordance with this subgenotypes in urban and suburban water samples (Fig. 4). After performing nested RT-PCR with HEV-specific finding, quantitative data on all tested samples of WWTPs indicate an HEV RNA reduction of about 1 log during primers, wastewater influent samples of two different WWTPs (WWTP 1 and WWTP 3) clearly showed the char- treatment. This result was validated by the HEV surveil- lance of WWTP 1 over a complete one-year period, showing acteristic 332 bp fragments. An exemplary agarose gel with amplified HEV nested RT-PCR products from HEV ORF1 an average decrease from 2 × 10 genome copies/100 ml in influent samples to 2 × 10 copies/100 ml in effluent samples is shown in Fig. 4a. Out of 173 tested wastewater influent samples, 15 samples (9%) displayed a clear band on the gel (Fig. 2). These effluent samples were taken before further UV treatment in the WWTPs. However, in summer, HEV and fragments were subjected to sequencing. None of the 1 3 144 Food and Environmental Virology (2020) 12:137–147 RNA reduction during wastewater treatment is expected to Detection and quantification of HEV RNA in environ- be higher, since WWTP 1 is run during the bathing season mental water samples is challenging. If low virus concentra- with an additional UV treatment of effluents prior to release tions are present in large sample volumes the methods used in surface waters. for virus concentration can have significant influences. We Elimination of viruses in WWTPs depend on the char- therefore compared the detection rates obtained by ultra- acteristic features of the viruses as well as on the structures centrifugation and PEG precipitation, using samples with or and combinations of treatment steps of the plants. Further- without virus concentration steps (Fig. 3). Direct sampling more, there is a lack of data for HEV RNA reduction during and ultracentrifugation revealed comparable monthly detec- treatment in WWTPs. So far, reports with quantified HEV tion rates, whereas the PEG-processed samples resulted in concentrations in environmental waters are rare and mainly lower HEV RNA findings. Direct virus detection is easy to restricted to wastewater influent samples. The HEV con- perform but since small volumes are used, the limit of detec- centrations in wastewater influents determined in this study tion is much higher than for methods with virus concentra- correspond to the concentrations reported by Rodriguez- tion steps. Due to the small tested volumes, calculated virus Manzano et al. (2010), Masclaux et al. (2013), Wang et al. concentrations could be over- or underestimated. Therefore, (2018b) and Miura et al. (2016). In contrast to the findings when samples are tested by different methods, final virus of Masclaux et al. (2013), where HEV RNA was detected concentration should always be reported together with the more frequently in summer, no clear seasonal pattern of concentration method to ensure valid comparisons of the HEV RNA occurrence was observed in the present study, obtained data. similar to the report of Ram et al. (2016). Ultracentrifugation and PEG precipitation are standard Since wastewater effluents are discharged into rivers, virus concentration methods for detection of HEV in sewage further investigations were carried out in two urban rivers samples (Clemente-Casares et al. 2003; Rodriguez-Manzano (Table 2). Under normal weather conditions, about 30% of et al. 2010; Masclaux et al. 2013; Myrmel et al. 2015; Ram these river samples were positive with low concentrations of et al. 2016; Iaconelli et al. 2017; Wang et al. 2018b; Matos HEV RNA. However, the median HEV RNA concentration et al. 2018). The 25% positive samples during the 1-year of all tested river samples was below the LOD. These low period using the PEG precipitation method were comparable detection rates and low RNA concentrations are reasonable to findings of Masclaux et al. (2013) and Miura et al. (2016), due to virus dilution in big water volumes. Similar results which reported rates of 32% and 22%, respectively. have been reported in Italian surface waters impacted by In our hands, detection rates were much higher using the runoffs from grazing land and discharges from treatment ultracentrifugation method. This method does not need any plants, where 25% of the tested water samples were HEV addition of chemicals and since viruses tend to attach to RNA-positive (Idolo et al. 2013). In line with the results suspended matter (Jin and Flury 2002), ultracentrifugation from the two rivers under normal conditions, only 15% of is a suitable method for influent samples of WWTPs. Moreo- samples from a bathing water were positive for HEV RNA, ver, these ultracentrifugated samples were most suited for with a median concentration of all tested samples below the genotyping of HEV strains. LOD. Therefore, no evidence of an increased health risk In contrast to quantitative RT-PCR detection, longer was found at this bathing area. This reflects the water man- fragments need to be amplified for genotyping, which may agement efforts to maintain the bathing water quality under result in a lower sensitivity. In the present study, 15 ampli- normal weather conditions. cons (9%) of the WWTP influent samples could be success- In urban areas, mixed channels for sewage and rain water fully sequenced. Using nested RT-PCR, similar results of may reach capacity limits after heavy rainfall events and 5–13.5% of influent samples positive for HEV RNA were lead to release of uncleared wastewater into rivers (com- reported from Italy and Spain (Rusiñol et al. 2015; Iaconelli bined sewer overflows, CSOs). et al. 2017; Alfonsi et al. 2018). For genotyping, samples Such CSOs seem to have a high impact on HEV detec- with virus concentrations steps were most suitable, since 13 tion rates and concentrations in rivers, as seen in this study amplicons were sequenced from ultracentrifugated or PEG- for river 1, where several CSO sites are located. After three precipitated samples. Although the nested primer system heavy rainfall events in summer 2016 causing CSOs in river was able to amplify all HEV genotypes (Johne et al. 2010), 1, HEV positive samples increased from 33 to 75%, with a only HEV-3 strains were detected in environmental water median copy number of 2 × 10 copies/100 ml. Therefore, samples. The most prevalent HEV subtype was HEV-3c. In urban rivers may contain high HEV RNA concentrations addition, two samples contained HEV genotype 3f. HEV- during rainfall-affected periods, thereby increasing the pub- 3c and HEV-3f were also recently reported in wastewater lic health risk of HEV infections over the faecal-oral route in Italy (Di Profio et al. 2019). Of our three samples which by bathing or recreational activities in the polluted urban could not be subtyped exactly, two are most likely of subtype rivers. 3a and one of subtype 3c or 3i. 1 3 Food and Environmental Virology (2020) 12:137–147 145 carefully timed representative samples of combined sewer overflow The detected environmental HEV genotypes correlate events. well with reported subgenotype data from clinical samples from Germany (Vollmer et al. 2012; Tabatabai et al. 2014; Open Access This article is licensed under a Creative Commons Attri- Adlhoch et al. 2016). Genotype HEV-3c was reported to bution 4.0 International License, which permits use, sharing, adapta- be the most prevalent genotype in German blood donors, tion, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, and genotypes 3a and 3e were also found in clinical sam- provide a link to the Creative Commons licence, and indicate if changes ples (Vollmer et al. 2012). Moreover, HEV-3c was identi- were made. The images or other third party material in this article are fied in the first German clinical report of acute hepatitis E included in the article’s Creative Commons licence, unless indicated during pregnancy (Tabatabai et al. 2014) and is the most otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not common type in the European Union/ European Economic permitted by statutory regulation or exceeds the permitted use, you will Area (EFSA 2017). need to obtain permission directly from the copyright holder. To view a Most of the wastewater influent samples were obtained from copy of this licence, visit http://creativ ecommons .or g/licenses/b y/4.0/. an urban WWTP with a catchment area of about 1.1 million people of the Berlin area. A recent study investigated HEV genotype 3 variants in patients from the same area and iden- References tified subtype 3c as the most prevalent HEV strain, besides genotypes 3e and 3f (Wang et al. 2018a). In our study, subtype Adlhoch, C., Avellon, A., Baylis, S. A., Ciccaglione, A. R., Couturier, E., de Sousa, R., et al. (2016). Hepatitis E virus: assessment of the HEV 3c was also detected in the suburban WWTP 3 with a epidemiological situation in humans in Europe, 2014/15. Journal pig farm located nearby. However, the overall HEV detection of Clinical Virology, 82, 9–16. rate did not differ from the other WWTPs. Besides domestic Alfonsi, V., Romanò, L., Ciccaglione, A. R., La Rosa, G., Bruni, R., pigs, in which HEV infection is highly prevalent (Fernández- Zanetti, A., et al. (2018). Hepatitis E in Italy: 5 years of national epidemiological, virological and environmental surveillance, 2012 Barredo et al. 2007; Jiménez de Oya et al. 2011; Dremsek et al. to 2016. Eurosurveillance, 23(41), 1700517. 2013), wild boars have been identified as a possible source Clemente-Casares, P., Pina, S., Buti, M., Jardi, R., Martín, M., Bofill- of HEV RNA. In addition, it has to be considered that Ber- Mas, S., et al. (2003). Hepatitis E Virus epidemiology in indus- lin also harbours a large population of wild boars (Stillfried trialized Countries. Emerging Infectious Diseases, 9(4), 449–454. Clemente-Casares, P., Ramos-Romero, C., Ramirez-Gonzalez, E., et al. 2017). In a current study, a high degree of nucleotide & Mas, A. (2016). Hepatitis E virus in industrialized countries: sequence homology in a wild boar isolate and a human isolate the silent threat. BioMed Research International. https ://doi. was detected and zoonotic HEV-3c and 3a were identified in org/10.1155/2016/98380 41. wild boars in the Berlin/Potsdam area (Schielke et al. 2009), Dalton, H. R., Kamar, N., & Izopet, J. (2014). Hepatitis E in developed countries: Current status and future perspectives. Future Microbi- corresponding with the detection of HEV-3c and 3a RNA in ology, 9(12), 1361–1372. this study. In cities, HEV transmission from wild boars to Di Profio, F., Melegari, I., Palombieri, A., Sarchese, V., Arbuatti, A., humans has to be taken into consideration, either by direct or Fruci, P., et al. (2019). High prevalence of hepatitis E virus in raw by indirect transmission by surface waters, environment, or sewage in Southern Italy. Virus Research, 272, 197710. Dremsek, P., Joel, S., Baechlein, C., Pavio, N., Schielke, A., Ziller, M., other carrier animals (Schielke et al. 2009). et al. (2013). Hepatitis E virus seroprevalence of domestic pigs The composition of subtypes detected in the environmen- in Germany determined by a novel in-house and two reference tal waters in the present study reflects the circulation of HEV ELISAs. Journal of Virological Methods, 190(1–2), 11–16. strains in humans and animals in the same region. Previous EFSA Panel on Biological Hazards (BIOHAZ), Ricci, A., Allende, A., Bolton, D., Chemaly, M., Davies, R., et al. (2017). Public health reports from other countries also demonstrated a correlation risks associated with hepatitis E virus (HEV) as a food-borne between detected viruses in wastewater and clinical cases pathogen. EFSA Journal, 15(7), e04886. (Hellmér et al. 2014; Ivanova et al. 2019). Faber, M., Willrich, N., Schemmerer, M., Rauh, C., Kuhnert, R., Stark, Detection of viruses in environmental waters can con- K., et al. (2018a). Hepatitis E virus seroprevalence, seroincidence and seroreversion in the German adult population. Journal of tribute to a better understanding of the epidemiology and Viral Hepatitis, 25(6), 752–758. prevalent strains in the population. Although foodborne Faber, M., Askar, M., & Stark, K. (2018b). Case-control study on risk transmission is considered as the main pathway of zoonotic factors for acute hepatitis E in Germany, 2012 to 2014. Eurosur- HEV infection, environmental transmission should not be veillance. https ://doi.org/10.2807/1560-7917.ES.2018.23.19.17- 00469 . neglected and water monitoring should be integrated in the Fenaux, H., Chassaing, M., Berger, S., Gantzer, C., Bertrand, I., & One Health approach to reduce public health risks. Schvoerer, E. (2019). Transmission of hepatitis E virus by water: An issue still pending in industrialized countries. Water Research, Acknowledgements Open Access funding provided by Projekt DEAL. 151, 144–157. S.B. was supported by a doctoral fellowship by the UBA. We thank F. Fernández-Barredo, S., Galiana, C., García, A., Gómez-Muñoz, M. T., Kubek (UBA) and the WWTP operating companies for wastewater Vega, S., Rodríguez-Iglesias, M. A., et al. (2007). Prevalence and samples. In addition, we are grateful to our partners in the BMBF genetic characterization of hepatitis E virus in paired samples of financed project Flusshygiene (02WRM1364B), especially W. Seis from the Berlin Centre of Competence for Water (KWB) for collecting 1 3 146 Food and Environmental Virology (2020) 12:137–147 feces and serum from naturally infected pigs. Canadian Journal Manor, Y., Blomqvist, S., Sofer, D., Alfandari, J., Halmut, T., Abra- of Veterinary Research, 71(3), 236. movitz, B., et al. (2007). Advanced environmental surveillance Garson, J. A., Ferns, R. B., Grant, P. R., Ijaz, S., Nastouli, E., Szypul- and molecular analyses indicate separate importations rather ska, R., et al. (2012). Minor groove binder modification of widely than endemic circulation of wild type 1 poliovirus in Gaza dis- used TaqMan probe for hepatitis E virus reduces risk of false trict in 2002. Applied and Environment Microbiology, 73(18), negative real-time PCR results. Journal of Virological Methods, 5954–5958. 186(1–2), 157–160. Masclaux, F. G., Hotz, P., Friedli, D., Savova-Bianchi, D., & Hellmér, M., Paxéus, N., Magnius, L., Enache, L., Arnholm, B., Oppliger, A. (2013). High occurrence of hepatitis E virus in Johansson, A., et al. (2014). Detection of pathogenic viruses in samples from wastewater treatment plants in Switzerland and sewage provided early warnings of hepatitis A virus and norovi- comparison with other enteric viruses. Water Research, 47(14), rus outbreaks. Applied and Environmental Microbiology, 80(21), 5101–5109. 6771–6781. Matos, A., Mesquita, J. R., Gonçalves, D., Abreu-Silva, J., Luxo, C., Iaconelli, M., Muscillo, M., Della Libera, S., Fratini, M., Meucci, L., & Nascimento, M. S. J. (2018). First detection and molecular De Ceglia, M., et al. (2017). One-year surveillance of human characterization of hepatitis E virus in water from wastewater enteric viruses in raw and treated wastewaters, downstream river treatment plants in Portugal. Annals of Agricultural and Envi- waters, and drinking waters. Food and Environmental Virology, ronmental Medicine, 25(2), 364–367. 9(1), 79–88. Meng, X. J. (2010). Hepatitis E virus: animal reservoirs and zoonotic Idolo, A., Serio, F., Lugoli, F., Grassi, T., Bagordo, F., Guido, M., risk. Veterinary Microbiology, 140(3–4), 256–265. et al. (2013). Identification of HEV in symptom-free migrants Miura, T., Lhomme, S., Le Saux, J. C., Le Mehaute, P., Guillois, and environmental samples in Italy. The Journal of Viral Hepa- Y., Couturier, E., et al. (2016). Detection of hepatitis E virus titis, 20, 438–443. in sewage after an outbreak on a French island. Food and Envi- Ivanova, O. E., Yarmolskaya, M. S., Eremeeva, T. P., Babkina, G. ronmental Virology, 8(3), 194–199. M., Baykova, O. Y., Akhmadishina, L. V., et al. (2019). Envi- Myrmel, M., Lange, H., & Rimstad, E. (2015). A 1-year quantitative ronmental surveillance for poliovirus and other enteroviruses: survey of noro-, adeno-, human boca-, and hepatitis E viruses in Long-term experience in Moscow, Russian Federation, 2004– raw and secondarily treated sewage from two plants in Norway. 2017. Viruses, 11(5), 424. Food and Environmental Virology, 7(3), 213–223. Jiménez de Oya, N., de Blas, I., Blázquez, A. B., Martín-Acebes, M. Pavio, N., Doceul, V., Bagdassarian, E., & Johne, R. (2017). Recent A., Halaihel, N., Gironés, O., et al. (2011). Widespread distribu- knowledge on hepatitis E virus in Suidae reservoirs and trans- tion of hepatitis E virus in Spanish pig herds. BMC Research mission routes to human. Veterinary Research, 48(1), 78. Notes, 4(1), 412. Pérez-Gracia, M. T., García, M., Suay, B., & Mateos-Lindemann, M. Jin, Y., & Flury, M. (2002). Fate and transport of viruses in porous L. (2015). Current knowledge on hepatitis E. Journal of Clinical media. In D. Sparks (Ed.), Advances in agronomy (Vol. 77, pp. and Translational Hepatology, 3(2), 117. 39–102). San Diego: Academic Press. Pina, S., Jofre, J., Emerson, S. U., Purcell, R. H., & Girones, R. Johne, R., Plenge-Boenig, A., Hess, M., Ulrich, R. G., Reetz, J., & (1998). Characterization of a strain of infectious hepatitis E Schielke, A. (2010). Detection of a novel hepatitis E-like virus virus isolated from sewage in an area where hepatitis E is not in faeces of wild rats using a nested broad-spectrum RT-PCR. endemic. Applied and Environment Microbiology, 64(11), Journal of General Virology, 91(3), 750–758. 4485–4488. Johne, R., Reetz, J., Ulrich, R. G., Machnowska, P., Sachsenröder, J., Ram, D., Manor, Y., Gozlan, Y., Schwartz, E., Ben-Ari, Z., Mendel- Nickel, P., et al. (2014). An ORF 1-rearranged hepatitis E virus son, E., et al. (2016). Hepatitis E virus genotype 3 in sewage derived from a chronically infected patient efficiently replicates and genotype 1 in acute hepatitis cases, Israel. The American in cell culture. Journal of Viral Hepatitis, 21(6), 447–456. Journal of Tropical Medicine and Hygiene, 95(1), 216–220. Jothikumar, N., Cromeans, T. L., Robertson, B. H., Meng, X. J., & RKI. (2019). Infektionsepidemiologisches Jahrbuch meldepflichtiger Hill, V. R. (2006). A broadly reactive one-step real-time RT- Krankheiten für 2018. Berlin, Germany: Robert Koch-Institut. PCR assay for rapid and sensitive detection of hepatitis E virus. Rodriguez-Manzano, J., Miagostovich, M., Hundesa, A., Clemente- Journal of Virological Methods, 131(1), 65–71. Casares, P., Carratala, A., Buti, M., et al. (2010). Analysis of Kimura, M. (1980). A simple method for estimating evolution- the evolution in the circulation of HAV and HEV in eastern ary rates of base substitutions through comparative studies of Spain by testing urban sewage samples. Journal of Water and nucleotide sequences. Journal of Molecular Evolution, 16(2), Health, 8(2), 346–354. 111–120. Rusiñol, M., Fernandez-Cassi, X., Timoneda, N., Carratalà, A., Krumbholz, A., Neubert, A., Joel, S., Girschick, H., Huppertz, H. I., Abril, J. F., Silvera, C., et  al. (2015). Evidence of viral dis- Kaiser, P., et al. (2014). Prevalence of hepatitis E virus antibod- semination and seasonality in a Mediterranean river catchment: ies in children in Germany. The Pediatric Infectious Disease implications for water pollution management. Journal of Envi- Journal, 33(3), 258–262. ronmental Management, 159, 58–67. Kumar, S., Stecher, G., & Tamura, K. (2016). MEGA7: molecular Rutjes, S. A., Lodder, W. J., Lodder-Verschoor, F., Van den Berg, H. evolutionary genetics analysis version 70 for bigger datasets. H., Vennema, H., Duizer, E., et al. (2009). Sources of hepatitis Molecular Biology and Evolution, 33(7), 1870–1874. E virus genotype 3 in The Netherlands. Emerging Infectious La Rosa, G., Fontana, S., Di Grazia, A., Iaconelli, M., Pourshaban, Diseases, 15(3), 381. M., & Muscillo, M. (2007). Molecular identification and genetic Schielke, A., Sachs, K., Lierz, M., Appel, B., Jansen, A., & Johne, R. analysis of norovirus genogroups I and II in water environ- (2009). Detection of hepatitis E virus in wild boars of rural and ments: comparative analysis of different reverse transcription- urban regions in Germany and whole genome characterization PCR assays. Applied and Environment Microbiology, 73(13), of an endemic strain. Virology Journal, 6(1), 58. 4152–4161. Smith, D. B., Paddy, J. O., & Simmonds, P. (2016a). The use of La Rosa, G., Pourshaban, M., Iaconelli, M., Vennarucci, V. S., & human sewage screening for community surveillance of hepa- Muscillo, M. (2010). Molecular detection of hepatitis E virus titis E virus in the UK. Journal of Medical Virology, 88(5), 915–918. in sewage samples. Applied and Environment Microbiology, 76(17), 5870–5873. 1 3 Food and Environmental Virology (2020) 12:137–147 147 Smith, D. B., Simmonds, P., Izopet, J., Oliveira-Filho, E. F., Ulrich, E virus infection in German blood donors. Journal of Clinical R., Johne, R., et al. (2016b). Proposed reference sequences for Microbiology, 50(8), 2708–2713. hepatitis E virus subtypes. The Journal of General Virology, Wang, B., Harms, D., Papp, C. P., Niendorf, S., Jacobsen, S., Lütge- 97(3), 537. hetmann, M., et al. (2018a). Comprehensive molecular approach Sridhar, S., Lau, S. K., & Woo, P. C. (2015). Hepatitis E: A disease for characterization of hepatitis E virus genotype 3 variants. of reemerging importance. Journal of the Formosan Medical Journal of Clinical Microbiology, 56(5), e01686–e1717. Association, 114(8), 681–690. Wang, H., Sikora, P., Rutgersson, C., Lindh, M., Brodin, T., Björle- Stillfried, M., Fickel, J., Börner, K., Wittstatt, U., Heddergott, M., nius, B., et al. (2018b). Differential removal of human patho- Ortmann, S., et al. (2017). Do cities represent sources, sinks genic viruses from sewage by conventional and ozone treat- or isolated islands for urban wild boar population structure? ments. International Journal of Hygiene and Environmental Journal of Applied Ecology, 54(1), 272–281. Health, 221(3), 479–488. Tabatabai, J., Wenzel, J. J., Soboletzki, M., Flux, C., Navid, M. H., & Schnitzler, P. (2014). First case report of an acute hepatitis E Publisher’s Note Springer Nature remains neutral with regard to subgenotype 3c infection during pregnancy in Germany. Journal jurisdictional claims in published maps and institutional affiliations. of Clinical Virology, 61(1), 170–172. Vollmer, T., Diekmann, J., Johne, R., Eberhardt, M., Knabbe, C., & Dreier, J. (2012). Novel approach for detection of hepatitis 1 3

Journal

Food and Environmental VirologySpringer Journals

Published: Jun 14, 2020

There are no references for this article.