Monitoring of Poyang lake water for sewage contamination using human enteric viruses as an indicator

Poyang Lake is the biggest fresh water lake in China and located in the north of Jiangxi Province, and to the middle and the lower reaches of the Yangtze River. This lake is connected with five main rivers including Gan River, Hu River, Xin River, Rao River, and Xiu River. Because its surroundings and natural resources, Poyang Lake provides local residents nearby opportunity for economic sources through fishing and the majority of drinking water. Therefore, the water quality of Poyang Lake is becoming a serious public health concern. In this study, water samples were taken from six locations along the Poyang Lake considering factors such as human sources and environmental impact (Fig. 1): the site Qingshanzha (1.8 million inhabitants nearby) located in Gan River which is selected since it is an important domestic wasterwater outlet (the rate of sewage treatment is 89%) from urban area in Nanchang; the site Guanniaotai and Tuoshan belong to Nanjishan Water of the southwest of Poyang Lake (total 3000 inhabitants), which are selected because of their kinds of biological resources especially migrant birds that may excrete feces into water; the site Xingzi (0.21 million inhabitants) and Wucheng (0.015 million inhabitants) are selected for their citizens live on Poyang Lake and visitors; and the site Dukou (0.29 million inhabitants), is the downstream of Poyang Lake, flows into Yangtze River, which may affected by upriver water quality. The site descriptions are summarized in Table 1. The water temperature of Poyang Lake was measured every sampling time for each of these six sites.

In this study, each site was monitored for four months from October of 2016 to January of 2017. Water samples were collected in the morning with sterilized 10-L polypropylene containers from the 10% hydrochloric acid (HCL) bath treatment as described previously [19, 20]. Samples were transported on ice to the laboratory of Nanchang University and processed for viral detection immediately. One Liter of water samples from each site was tested for HEV and another liter was for fecal indicator bacteria, respectively

Sample concentration, nucleic acid isolation, and RT-PCR were performed using the protocols established previously in this laboratory that showed high efficiency of viral recovery from environmental water samples [19, 21‐23]. In brief, 5 ml of 5 mol/L Magnesium Chloride was added to one liter of freshwater sample (final concentration of 25 mM) at the laboratory, and the sample was allowed to set for 5 min at room temperature before filtration. Samples were concentrated using negatively-charged type HA filter membranes (EMD Millipore, Billerica, MA, USA) with a 0.45 μm pore size and 47 mm in diameter under vacuum pump filtration, in a filtration-based method described previously [19]. One-liter water sample was filtered through for each fresh water sample, except when filters clog, filtration could take up to more than one hour, and the final amount of filtrated water was recorded. Nucleic acids were extracted from the recovered membranes using the Power Water® RNA Isolation Kit (MoBio Laboratories, Inc., Carlsbad, CA, USA), according to a modified protocol [20]. The final volume for each sample extraction was 60 μL, of which an aliquot 15 eluent was stored at −80 °C for viral DNA assays. The remaining 45 μL was used for RNA extraction by removing residual DNA through treating the extract with 5 μL of 10X DNase I buffer and 4 μL of DNase I. Following incubation at RT for 15–20 min. 0.75 μL of 0.5 M EDTA was added to the extract to protect RNA when DNase I was heated inactivate at 75 °C for 5 min. The recovered RNA was kept at −20 °C until use. RNA was subject to RT-PCR using the Easy Script® One-Step gDNA Removal and cDNA Synthesis Super Mix (Transgene, China). Reverse transcriptase PCR (RT-PCR) was used previous standardization and optimization as follows: 1 μL of Random Primer, 10 μL of 2 × ES Reaction Mix and 1 μL of Easy Script® RT/RI Enzyme Mix were added to 8 μL of sample RNA. Then cDNA samples were stored at 4 °C until use.

PCR was conducted using the protocol described previously [19, 21] and tested for the following four enteric viruses: norovirus geno-groups I and II, enterovirus, and adenovirus. PCR conditions and primer set for each virus detection are summarized in Table 2. Both negative and positive controls were included for each PCR performance. In particular, deionized water and viral plasmid DNA prepared previously for each of the four viruses were used as negative and positive controls, respectively. The first PCR product was used as template for the second amplification under the same PCR condition. Results were visualized using gel electrophoresis on a 2% agarose gel stained with ethidium bromide.

To verify accurate DNA amplification and viral detection, 51 PCR products from the nest PCR were cloned using TA Cloning kit purchased from Invitrogeen and sent to the Beijing Genomics Institute (BGI) for two directional sequencing. The sequencing data were jointed and compared to the available nucleic acid sequences in the National Center for Biotechnology Information (NCBI) database using Basic Local Alignment Search Tool (BLAST).

The bacteriological analyses were done according to the filter membrane method of GB/T 5750.12–2006 《Standard Examination Methods for Drinking Water》of China. The detected indicators including Aerobic bacterial Count (ACC), total coliform (TC), fecal coliform (FC) and E. coli. One milliliter of different dilution water were passed through 0.45 μM pore size, 39 mm diameter Millipore membrane filters, which were placed onto different culture media: ACC: Nutrient Agar, incubated at 36 °C for 48 h; TC: Endo Agar, incubated at 36 °C for 18–24 h; FC: MFC medium, incubated at 44.5 °C for 24 h; and E. coli: MUG medium, incubated at 44.5 °C for 24 h.

According to the following two standards to evaluating the water functions: (1). The average TC of source water in GJ3020–1993 《Standard Examination Methods for Drinking Water》 is 1000 CFU/L for I level water, 10,000 CFU/L for II level water. (2). GB3838–2002《Standards of surface water quality》for FC: I to IV level water quality limit is 200, 2000, 1 × 10⁴, 2 × 10⁴, 4 × 10⁴/L.

Statistical analysis was carried out using SPSS version 19.0. The Spearman rank correlation was used to test the relationship between bacterial indicators and enteric viruses. P-value below 0.05 was considered statistically significant, and all p-values were given as two-tailed.

A total of twenty-four water samples were collected from six different sites along the Poyang Lake and analyzed. All sample sites tested positive for more than two enteric viruses during the 4-month study period. All these 4 HEV were tested positive from all the 6 sites at least once (Fig. 2). Nineteen of twenty-four samples (95%) tested positive for two or more viruses. As shown in Table 3, human enterovirus (58%) and adenovirus (67%) were more frequently detected from these six sites, followed by were norovirus I (50%) and norovirus II (38%). Sixteen of twenty-four water samples were tested positive for adenovirus, followed by enterovirus (14/24) and norovirus geno-group I (12/24). Under the described laboratory conditions, human norovirus geno-group II was tested positive in 9 of the 24 samples representing the least commonly detected virus in this study.

All the selected PCR amplicons were sequenced viral detection verification and confirmed these positive detections were correct amplification as the viruses of interest. Several viral subtypes of enteric viruses were identified in Poyang Lake including Norovirus strains of GI.9, GI.7, GI.Toyama, GII.NEC, GII.Seoul, GII.CC, GII.Hu, and GII.CC, and human adenovirus type 41. Several other human enteric viruses were also detected in this study including human echovirus type 11, 25, E29 and human coxsackievirus B3. Full sequencing results were summarized in Table 4.

The results of indicator bacteria are expressed as colony forming units (CFU) per 1 ML (Tables 5 and 6). The results show that TC varied between 0 and 2.2 × 10³ CFU/mL in November, and FC levels were between 0 and 2.4 × 10⁵ CFU/mL in November. Twenty-three of the 24 water samples tested met the TC I level of CJ3020–1993 water quality criteria. And only 1 water sample tested for FC did not meet I level water quality of GB3838–2002 criterion. Six sites were all met I level water quality of GB3838–2002 criterion in microbiological level, but these sites all tested positive for enteric viruses. The detection of majority of enteric viruses did not show significant correlation with the detection of bacteria indicators (Table 6).