International Journal of Hygiene and Environmental Health 213 (2010) 210–216

Contents lists available at ScienceDirect

International Journal of Hygiene and Environmental Health journal homepage: www.elsevier.de/ijheh

Chemical and microbiological parameters as possible indicators for human enteric viruses in surface water Lars Jurzik a,∗,1 , Ibrahim Ahmed Hamza a,1,2 , Wilfried Puchert b , Klaus Überla c , Michael Wilhelm a a b c

Department of Hygiene, Social and Environmental Medicine, Ruhr-University Bochum, Germany Health Department of the State of Mecklenburg-West Pomerania, Schwerin, Germany Department of Molecular and Medical Virology, Ruhr-University Bochum, Germany

a r t i c l e

i n f o

Article history: Received 19 February 2010 Received in revised form 7 May 2010 Accepted 7 May 2010 Keywords: Surface water Enteric viruses Somatic coliphages TCPP Indicator

a b s t r a c t There are still conflicting results on the suitability of chemical and microbiological parameters as indicators for the viral contamination of surface waters. In this study, conducted over 20 months, the abundance of human adenovirus, human polyomavirus, enterovirus, group A rotavirus and norovirus was determined in Ruhr and Rhine rivers, Germany. Additionally, prevalence of different possible indicators such as somatic coliphages, E. coli, intestinal enterococci, and total coliforms was also considered. Moreover, the chemical parameter TCPP (tris-(2-chloro-, 1-methyl-ethyl)-phosphate), characterized by environmental stability and human origin, was included. Furthermore, chemical parameters (fluoride, chloride, nitrate, nitrite, bromide, phosphate, and sulfate) which may influence the stability and subsequently the detection rates of viruses in aquatic environment were measured. Quantitative Real-Time (RT-)PCR and double agar layer test were used for the quantification of human enteric viruses and somatic coliphages, respectively. The analyses for E. coli, total coliforms, and intestinal enterococci were done with respect to the standard reference method. The chemical parameters were measured by liquid chromatography of ions and by gas chromatography-flame photometer detector (GC-FPD), respectively. We demonstrated that human adenovirus had the highest detection rate (96.3%), followed by somatic coliphages (73.5%), human polyomavirus (68.6%), and rotavirus (63.5%). However, norovirus GII and enterovirus were found in only 25.7 and 17.8%, respectively. The concentration of the viral genome ranged between 16 and 1.1 × 106 gen. equ./l (genome equivalents/l) whereas the concentrations for TCPP ranged between 0.01 and 0.9 ␮g/l. The results of the Pearson correlation showed no association between TCPP and any other microbiological parameter. None of the other tested chemical parameters correlated negatively, and therefore they do not influence the stability of enteric viruses. We conclude that neither TCPP nor any other chemical or microbiological parameter can be used as a reliable indicator for the presence of enteric viruses in river water. © 2010 Published by Elsevier GmbH.

Introduction Several years ago more than 100 species of viruses have been isolated from sewage water (Cukor and Blacklow, 1984). It is well known that pathogenic microorganisms may enter surface waters through discharges of raw and treated sewage, and also manure runoff from agricultural land (Pusch et al., 2005; van den Berg et al., 2005). Some enteric viruses have been related to waterborne outbreaks by a fecal contamination of drinking water. This contamination may be caused by sewage or surface water, breakdown

∗ Corresponding author. Tel.: +49 234 32 28931; fax: +49 234 32 14199. E-mail address: [email protected] (L. Jurzik). 1 These authors contributed equally to this work. 2 Permanent address: Environmental Virology Laboratory, Department of Water Pollution Research, National Research Centre, 12311 Dokki, Cairo, Egypt. 1438-4639/$ – see front matter © 2010 Published by Elsevier GmbH. doi:10.1016/j.ijheh.2010.05.005

of the drinking water supply, a bypass connection to an irrigation system or leakages in the water distribution system (Hafliger et al., 2000). Some waterborne outbreaks with enteric viruses have been described in the last years. One of them took place in southern Italy with 344 norovirus (NoV) infected people (Boccia et al., 2002). Additionally, 460 people were infected with enterovirus (EV) during an outbreak in Belarus (Amvrosieva et al., 2001). Furthermore, a recent waterborne norovirus outbreak occurred in Podgorica (Montenegro) with 1699 people infected due to contaminated municipal water (Werber et al., 2009). The main problem with determining enteric viruses in water directly is that the detection methods like cell culture and Real-Time PCR, or a combination of both, are expensive, timeconsuming, and labor-intensive. This leads to the concept of indicator microorganisms. Parameters like somatic coliphages, E. coli, total coliforms, and intestinal enterococci are often discussed as indicators for enteric

L. Jurzik et al. / International Journal of Hygiene and Environmental Health 213 (2010) 210–216

211

Fig. 1. Location of sampling sites, sewage treatment plants and direct discharge. Samples were taken biweekly with a volume of 10 l.

viruses in surface water. Somatic coliphages use E. coli and closely related species as hosts and hence can be released by these bacterial hosts into the feces of humans and other warm-blooded animals. Coliphages used in water quality assessment are divided into the major groups of somatic coliphages and F-specific RNA (F-RNA) phages, which are most frequently studied (Skraber et al., 2004a). Phages share many properties with human viruses, notably composition, morphology, and structure. Due to their resistance against environmental factors, somatic coliphages are more applicable than fecal bacteria for indicating fecal contamination of water (Contreras-Coll et al., 2002). In the present study, an additional chemical parameter, the organophosphate TCPP, Tris (2-chloro-, 1-methyl-ethyl)phosphate, has been included to find a suitable indicator for enteric viruses in surface waters. As could be shown in a human biomonitoring study TCPP is absorbed and distributed to the whole body and the metabolites are excreted via urine (Schindler et al., 2009). Due to its high environmental stability it passes the sewage treatment plant without reduction and afterwards enters the surface water (Andresen et al., 2004; Fries and Puttmann, 2001). Supportive to the human biomonitoring study a high correlation (R2 = 0.92) between the TCPP charge of the outlet of sewage treatment plants and the associated population has been demonstrated (MUNLV, 2008). For this reasons we tried to evaluate its suitability as indicator for contamination of river water with human enteric viruses. In addition to the microbiological parameters and TCPP, this paper also considers those factors which may influence the stability and the survival of viruses in river water. Presently, there are no suitable models developed that can reliably predict the presence of enteric viruses in surface water, and there is less concordance between the published data concerning the suitability of some viruses/coliphages as indicators for enteric viruses in water. In our recent study we detected at the Ruhr and Rhine rivers (water supply of about 5.3 million people) nucleic acids of several human pathogenic viruses during winter 2007/08 (Hamza et al., 2009). In this study water samples have been tested

over a period of 20 months to determine the relation between human viruses and some chemical and microbiological parameters in surface waters in Germany. Methods Study area The water samples were collected from five sampling sites located along the river Ruhr, which meets the Rhine at Duisburg (North Rhine-Westphalia, NRW), Germany. In detail, water samples were collected near to Hagen, Bochum, Essen, Muelheim (all situated along the river Ruhr in the given order), and Duesseldorf (river Rhine, 30 km upstream of Duisburg). The distances to the next upstream sewage treatment plant is 1.5–10 km. The location of sampling sites is shown in Fig. 1. Sampling and virus concentration Water samples (10 l each) were collected from February 2008 to September 2009, in a depth of minimum 0.3 m, to avoid disinfectant effect of the UV light (Hijnen et al., 2006). Water temperature and the pH value were measured in situ. The samples were transported within the next 4 h to the laboratory and were subsequently analyzed. The concentration of viruses and coliphages by filtration procedure (VIRADEL) has already been described before, followed by PEG-6000 precipitation as a reconcentration step (Hamza et al., 2009). In brief MgCl2 was added to the sample with a final concentration of 0.05 M. Additionally, the pH was adjusted at 3.5 with HCl (APHA, 1998). A negatively charged membrane with a pore size of 0.45 ␮m (Millipore, Bedford, USA) and a diameter of 142 mm was used. After filtration the membrane was rinsed with 200 ml of 0.5 mM H2 SO4 with a pH of 3.5. The viruses were eluted with 70 ml of a buffer containing 0.05 M KH2 PO4 , 1 M NaCl, 0.1% Triton X-100 with a pH of 9.2. The eluate was neutralized using 1N NaOH. After-

212

L. Jurzik et al. / International Journal of Hygiene and Environmental Health 213 (2010) 210–216

wards it was reconcentrated by adding 12.5% polyethylene glycol (PEG-6000, Merck, Darmstadt, Germany) and 2.5% NaCl. The eluate was stirred at 4 ◦ C for 4 h, centrifuged at 10,000 × g for 30 min and pellet was suspended in 3 ml PBS.

Table 1 Detection rate of microbiological parameters measured during our study. For HAdV, E. coli, total coliforms and intestinal enterococci more than 90% of all collected samples were positive. Parameter

N

No. (%) of positive samples

No. (%) of negative samples

HAdV HPyV NoV GII EV RoV Somatic coliphages E. coli Total coliforms i enterococci

190 188 187 174 181 185 182 185 192

183 (96.3) 129 (68.6) 48 (25.7) 31 (17.8) 115 (63.5) 136 (73.5) 150 (82.4) 179 (96.8) 175 (91.1)

7 (3.7) 59 (31.4) 139 (74.3) 143 (82.2) 66 (36.5) 49 (26.5) 32 (17.6) 6 (3.2) 21 (10.9)

Quantitative Real-Time PCR Viral nucleic acids were extracted from 200 ␮l of the concentrated virus suspension using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. The Real-Time (RT-) PCR was performed as previously described (Hamza et al., 2009), except Taqman probe was used instead of sybr green protocol for human adenovirus (HAdV) qPCR to increase the specificity to human adenovirus strains (Heim et al., 2003). In brief, the PCR was carried out using Quantitect probe RT-PCR kit (Qiagen, Hilden, Germany). The amplification conditions were optimized for each virus group and were as follows: (i) HAdV; 95 ◦ C for 15 min; 45 cycles of 95 ◦ C for 15 s, 60 ◦ C for 60 s, (ii) RoV; 50 ◦ C for 30 min, 95 ◦ C for 15 min, 45 cycles of 95 ◦ C for 15 s, 56 ◦ C for 30 s, 72 ◦ C for 1 min. (iii) NoV GII; 50 ◦ C for 30 min, 95 ◦ C for 15 min, 10 cycles of 95 ◦ C for 15 s, 56 ◦ C for 1 min, 35 cycles of 95 ◦ C for 15 s, 63 ◦ C for 1 min. (iv) HPyV; 95 ◦ C for 15 min, 50 cycles of 95 ◦ C for 15 s, 60 ◦ C for 1 min. (v) EV; 50 ◦ C for 30 min, 95 ◦ C for 15 min, 45 cycles of 95 ◦ C for 15 s, 60 ◦ C for 1 min. The specificity of the PCR assay was tested previously by sequence analysis (Hamza et al., 2009). Detection limit of the assay and PCR inhibition control was tested as has been published before by Hamza et al. (2009).

HAdV – adenovirus; EV – enterovirus; NoV GII – norovirus GII; RoV – rotavirus; HPyV – human polyomavirus; i enterococci – intestinal enterococci.

Statistical analysis Statistical analysis was performed to investigate the association between microbiological and viral load in surface waters. For the chemical parameters the data below the detection limit were set to 1/2 of the detection limit. Statistica software (9.0) was used to calculate the Pearsons correlation. To evaluate the concentration of the detected parameters at the different sampling site ANOVA test was performed. Results Detection of viral nucleic acids and bacteria

Quantification of somatic coliphages Somatic coliphages were quantified using the double layer plaque assay according to the method of the International Organization for Standardization (ISO, 2002). For surface water the E. coli strain DSM 13127 was used. Bacteriological analysis The bacteriological analyses were done according to the method of the International Organization for Standardization (ISO, 2001). For E. coli, total coliforms, and enterococci detection: 100 ml of water was passed through 0.45 ␮m membrane filters (47 mm diameter, Millipore, Bedford, USA), which were subsequently placed onto different culture media (Oxoid, Cambridge, UK). E. coli and total coliforms: lactose utilization on lactose-TTC agar, then oxidase- and indol-testing. Intestinal enterococci: a two step selective agar method with Slanetz-Bartley- and Aesculin agar (ISO, 2000). Chemical analysis The chemical analysis of dissolved bromate and chloride, fluoride, nitrate, nitrite, phosphate and sulfate was conducted according to the method of the International Organization for Standardization (DIN, 2001, 2008) and was done by ICS-90 liquid chromatography of ions (Dionex, USA). TCPP in surface water was analyzed by gas chromatography-flame photometer detector (GC-FPD) after solid phase extraction (SPE). The method has been described previously by Prösch et al. (2002). The detection limits for the chemical parameters were: 0.1 mg/l for fluoride, 0.002 mg/l for bromate, 1.0 mg/l for chloride, 0.02 mg/l for nitrite, 1.0 mg/l for nitrate, 0.1 mg/l for phosphate, 1.0 mg/l for sulfate, and 0.02 ␮g/l for TCPP.

Water samples were taken at five different sampling sites along the rivers Ruhr and Rhine over a period of 20 months and were concentrated using the VIRADEL method. For each parameter a statistical test was performed to clarify whether there are significant differences for the concentration of the detected parameters between the five sampling sites. Only for somatic coliphages a significant difference was found (data not shown). Table 1 represents detection rate of microbiological parameters measured during the study, whereas the descriptive statistic for the viral parameters is presented in Table 2. HAdV (human adenovirus, n = 190) showed the highest detection rate (96.3%) with a median concentration of 2.9 × 103 gen. equ./l and a maximum of 7.3 × 105 gen. equ./l. While the detection rates of HPyV (human polyomavirus, n = 188), somatic coliphages (n = 185), and RoV (rotavirus, n = 181) were 68.6, 73.5, and 63.5%, respectively. NoV GII (norovirus GII, n = 187) and EV (enterovirus, n = 174) have been detected in only 25.7 and 17.8% of the examined samples. All samples were negative for norovirus GI. Due to the fact that temperature is one of the main factors influencing the stability of enteric viruses in surface water the correlation with other microbiological parameters has been conducted with different temperatures (i) ≥10 ◦ C and (ii)