- Email: [email protected]
Human adenoviruses as waterborne index pathogens and their use for Quantitative Microbial Risk Assessment
Science of the Total Environment 651 (2019) 1469–1475
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Human adenoviruses as waterborne index pathogens and their use for Quantitative Microbial Risk Assessment Marco Verani a, Ileana Federigi a,⁎, Gabriele Donzelli a, Lorenzo Cioni b, Annalaura Carducci a a b
Laboratory of Hygiene and Environmental Virology, Department of Biology, University of Pisa, Via S. Zeno 37, 56127 Pisa, Italy Scuola Normale Superiore, P.zza dei Cavalieri, 7, 56126 Pisa, Italy
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• HAdV quantification in different water samples in parallel with fecal indicators • 64% of samples positive for HAdV, decreasing from WWTP (100%) to seawaters (21%) • HAdV occurrence in seawater samples complied with excellent EU bathing water criteria • Significant correlation between HAdV and somatic coliphages in less polluted waters • HAdV to indicators ratios modelled as probability density functions for QMRA
a r t i c l e
i n f o
Article history: Received 31 May 2018 Received in revised form 22 September 2018 Accepted 22 September 2018 Available online 25 September 2018 Editor: Paola Verlicchi Keywords: Human adenovirus Escherichia coli Intestinal enterococci Somatic coliphages QMRA Water safety
a b s t r a c t The current microbial water quality standards are based on the monitoring of fecal indicator organisms, which are mainly bacterial indicators (i.e., Escherichia coli, intestinal enterococci), however epidemiological data indicate that viruses are important etiological agents of waterborne illnesses. Among waterborne viruses, human adenovirus can be considered as an index pathogen, owing to its abundance in sewage and persistence in the environment, as well as its potential infectivity. In this study, data on human adenoviruses from different water matrices (the entrance and exit of a water treatment plant, rivers and seawaters) were analyzed, in parallel with traditional fecal bacterial indicators and somatic coliphages. The results showed a 64% frequency of positive adenovirus samples, decreasing from the sewage system (100% at the entrance and 94% at the exit) to rivers (92% and 72% for different rivers) and seawater (21%). Adenovirus concentrations showed a significant correlation with somatic coliphages in one river and seawater, thus supporting the recent inclusion of coliphages as viral indicators in water safety guidelines. The data collected were used to estimate adenovirus to indicator ratios, which could be used as input in Quantitative Microbial Risk Assessment (QMRA) studies. © 2018 Published by Elsevier B.V.
1. Introduction ⁎ Corresponding author. E-mail addresses: [email protected] (M. Verani), [email protected] (I. Federigi), [email protected] (G. Donzelli), [email protected] (L. Cioni), [email protected] (A. Carducci).
https://doi.org/10.1016/j.scitotenv.2018.09.295 0048-9697/© 2018 Published by Elsevier B.V.
The fecal pollution of waters used for different purposes (potable, reuse or recreation) is a public health issue. The consumption of, or contact with, contaminated waters is associated with outbreaks caused by
1470
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
waterborne pathogens, above all enteric viruses and protozoan parasites (Love et al., 2014). Of the enteric viruses, special attention has been given to the human adenovirus (HAdV) owing to its diffusion, epidemiological importance and easy detection. HAdV is a member of the genus Mastadenovirus in the Adenoviridae family, which comprises at least 51 serotypes, which are classified into seven species (A–G). HAdV is responsible for a wide range of health problems (including respiratory, gastrointestinal, and urinary infections), but also for a large number of asymptomatic infections, and all HAdV types can be excreted in high concentration into feces from infected individuals (Silva et al., 2011). These features have justified the inclusion of HAdVs in the US Environmental Protection Agency's water contaminant candidate list (U.S. EPA, 2016). Moreover, many studies have also suggested using HAdV as an indicator of the viral contamination in water (Wyer et al., 2012; Hewitt et al., 2013), or of the efficiency of viral removal in wastewater depuration plants (Carducci et al., 2008; La Rosa et al., 2010). These recommendations regarding HAdVs are made in relation to several specific qualities: - their high frequency retrieval and persistence in different water matrices such as wastewater effluents, rivers, seawaters and drinking waters (Haramoto et al., 2007; Wyn-Jones et al., 2011; Fongaro et al., 2013); - their resistance to water treatment/disinfection highlighted by studies that have revealed the presence of viable particles in waters treated by chlorine (van Heerden et al., 2005; Thurston-Enriquez et al., 2005) and their resistance to ultraviolet (UV) disinfection (Linden et al., 2007); - their higher frequency and abundance in waters compared to other enteric viruses, with genomic copies (GC) concentrations ranging from ~108 GC/100 ml in urban sewage (La Rosa et al., 2010; Carducci and Verani, 2013) to ~103 GC/100 ml in surface waters (Poma et al., 2012; Vieira et al., 2016); - their association with disease outbreaks in recreational waters (Sinclair et al., 2009); - the stability of their genome in comparison to RNA viruses (i.e. norovirus) guaranteeing a more “conservative” approach for risk assessment; - the possibility of cultivation (for most of the strains) and of biomolecular detection which enables the proportion of infective particles on genome units to be estimated. HAdV is thus considered as a promising indicator of viral pollution in different water-related area (Albinana-Gimenez et al., 2009; Hewitt et al., 2013) and is used as an index pathogen in Quantitative Microbial Risk Assessment (QMRA) in many different settings such as natural and treated recreational waters (Kundu et al., 2013), reused waters (Verbyla et al., 2016) and bioaerosol from wastewater treatment plants (Carducci et al., 2018). Nevertheless, current microbial water quality standards are mainly based on the monitoring of fecal indicator bacteria because of the low complexity and inexpensive detection methods. These microorganisms represent indicators of fecal contamination from warm-blooded animals (total and fecal coliform, E. coli, intestinal enterococci, Clostridium perfringens) some of which are specific indicators of human fecal pollution (E. coli and intestinal enterococci) (WHO, 2011). In addition, alternative microorganisms have been proposed as indicators used to assess viral water safety, such as somatic coliphage, which were added to the list of indicators in recent guidelines for recreational waters (U.S. EPA, 2018) and in the proposed EU regulation for drinking waters safety (EU, 2018). In the context of public health protection, these parameters are still an issue of great controversy, because of the lack of consistent correlation between bacterial indicators and enteric viral pathogens, such as the Adenoviridae family (Girones and Bofill-Mas, 2013; Love et al., 2014) and the reliability of somatic coliphage used as viral indicators
which is still under discussion (Lin and Ganesh, 2013). Considering the possible use of HAdV as both an indicator of viral pollution and index pathogen for QMRA, in this study we evaluated its relationship to other indicators. We thus analyzed the results of the study of different types of water (sewage, sea and rivers) conducted by the Hygiene and Environmental Virology Laboratory of the University of Pisa. The aim was to compare the occurrence and the amount of HAdV with those of fecal bacteria indicators - FIB (E. coli - EC, and Intestinal Enterococci IE) and viral indicators (Somatic Coliphages - SC). We also used the collected data to determine the ratios (point estimates and modelled as probability density functions) between HAdV and EC, IE, and SC, to be used for QMRA purposes, in order to estimate the HAdV concentrations on the basis of the indicator's concentrations. This approach was included by WHO (2016) in the current guidelines for the QMRA application to Water Safety Management. 2. Material and methods 2.1. Samples During several environmental monitoring campaigns, that were part of the daily sampling activities of our laboratory, we analyzed 241 water samples that were collected in central-northern Italy. All samples were collected in sterile containers, transported to the laboratory in cooler boxes containing ice and analyzed within 24 h. In particular: 1. 104 samples (52 of which were from the plant entry and 52 from the plant exit) were taken from a municipal wastewater treatment plant (WWTP), situated in Pisa; 2. 100 seawater samples were derived from monitoring campaigns on the Tyrrhenian coast, from bathing waters located at river mouths; 3. 37 samples were taken from 2 rivers of which 26 were from a small river that runs through an alluvial plain in northern Tuscany impacted by urban sewage and diffuse pollution sources (River No 1), and 11 were taken from the river Pescia, a tributary of the river Arno (River No 2) affected above all by agricultural runoff. For HAdV detection, 10 L of each water sample were analyzed except for the wastewater treatment effluent, for which the sampling volume was 1 L. For the determination of EC, IE, and SC, a sample volume of 100 ml was analyzed. 2.2. Samples treatment for HAdV detection Samples were concentrated to 40 ml volume, using a two-step tangential-flow ultrafiltration already previously described, with a recovery efficiency up to 100% (Carducci et al., 2003; La Rosa and Muscillo, 2013). Briefly, the samples were filtered in two steps using two devices: an apparatus equipped with polysulfone membrane with a 10,000 MW exclusion size to give a final 250 ml sample eluted in 3% beef extract at pH 9, then the second concentration was performed with a Mini-ultrasette apparatus, to obtain the concentrated sample. After a pH neutralization, the concentrated samples were treated with chloroform to eliminate bacteria and enveloped viruses (Carducci et al., 2008). 2.3. Biomolecular analysis Viral DNA was extracted from concentrated samples using a QIAamp DNA kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The extracted nucleic acid was assayed by real time quantitative PCR (qPCR) (Bofill-Mas et al., 2006). All reactions were performed in triplicate in 96-well optical plates using an ABI 7300 sequence detector system (Applied Biosystems, Foster City, California) and, for quantification, viral GC numbers were obtained from the standard curve dilution
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
series equation (range = 102–109). The standard curve was constructed by cloning the entire hexon region of Ad41 into pBR322 (Verani et al., 2016). In each plate, a positive control and a negative control were tested separately. A control for the presence of enzymatic inhibitors was also performed by analysing both undiluted and 10-fold diluted extracted DNA for each sample (La Rosa and Muscillo, 2013). To determine if the negative signals in the diluted extracts were due to low DNA amount, these DNA extracts were spiked with known concentration of plasmid DNA (103 copies/reaction) (internal control). An increase in the threshold cycles (Ct) of spiked DNA extracts was considered to indicate qPCR inhibition (Hamza et al., 2009). The detection limit for quantitative PCR was ~1 GC/reaction. 2.4. Fecal indicators Fecal indicator bacteria were analyzed according to ISO 9308-3 (1998) for EC, and ISO 7899-1 (1999) for IE. Cell density was enumerated using a microplate method, and counted as the most probable number (MPN)/100 ml of sample. SC titres were determined following the ISO 10705-2 (2001) cultural method based on double agar layer plaque assay using E. coli as host strain (E. coli strain C (ATCC 13706) for river and seawaters and E. coli strain CN (ATCC 700078) for sewage samples). The results were counted as the Plaque Forming Unit (PFU)/ 100 ml of sample. 2.5. Data analysis The microbial concentrations data were Log10 transformed and the geometrical means were calculated. MedCalc Statistical Software 14 (MedCalc Software, Ostend, Belgium) was used for multiple regression analyses in order to assess the potential correlations between the HAdV concentrations and the microbial indicators (EC, IE, and SC) in each type of water. For the regression analyses, a ‘viral load’ corresponding to half of the detection limit was attributed to samples with a negative qPCR (Nordgren et al., 2009). Boxplot graphs were generated with Excel for Windows (Microsoft Office Excel 2016, Redmond, Washington, USA). Each graph represents the whiskers of the minimum and maximum values and the boxes from the 1° and 3° quartile of the data. To compare HAdV occurrence in waters with different levels of bacterial pollution, river and seawater samples were divided based on the threshold for “excellent” water quality according to the EU bathing water directive, with bacterial concentrations expressed in Colony Forming Units (CFU)/100 ml (EU, 2006): 100 CFU/100 ml for IE and 250 CFU/100 ml for EC in river waters, and 200 CFU/100 ml for IE and 500 CFU/100 ml for EC in seawater. A chi square test was then
1471
performed using MedCalc software to compare the frequency of HAdV positivity in waters above the bacterial threshold with the ones below. 2.6. Estimation of ratios between HAdV and fecal indicators Only samples with quantifiable concentrations of at least one parameter were considered for the ratios calculation. The ratios between GC of HAdV and indicators in each type of water were estimated through a best-fit probability distribution function, selected using EasyFit software (MathWave Technologies, accessed at http:// www.mathwave.com/en/home.html on May 2018) according to the Kolmogorov-Smirnov (KS) test as a goodness of fit statistical test. 3. Results The overall frequency of HAdV positive samples was 64%, with a significant decreasing trend from untreated sewages to the seawater: 100% in WWTP entry, 94% in WWTP exit, 92% in River No 1, 72% in River No 2 and 21% in seawater (p b 0.001). A similar trend was found among the geometric means of the concentrations: 8.78 ± 1.19 Log GC/l for WWTP entry, 6.70 ± 1.77 Log GC/l for WWTP exit, 5.66 ± 1.59 Log GC/l for River No 1, 3.33 ± 1.00 Log GC/l for River No 2 and 2.56 ± 0.62 for seawater. These different concentrations are in accordance with the progressive dilution effect of the fecal contamination from sewages to the cleaner surface waters. The control for the presence of enzymatic inhibitors showed no inhibitory effects for DNA amplification in qPCR. For the FIB and SC concentrations, the observed trend among rivers was different, because the microbial pollution in River No 2 was higher than in River No 1 (Fig. 1). In seawater, all samples were below the threshold for “excellent” water quality (EU, 2006) based on FIB (see Section 2.5 Data Analysis), however they were positive for HAdV (21%). In River No 1 and River No 2, 77.8% (17/22) and 75% (6/8) of samples, respectively, were positive for adenovirus, and they exhibited FIB contamination under the threshold. Both in rivers and seawaters, the HAdV frequency was not correlated to water quality based on FIB (chi square, p = 0.92 for River No1 and p = 0.20 for River No2, p = 0.32 for seawater). The results of multiple regression analysis showed a significant relationship among HAdV and indicators (p b 0.05) in River No 2 and seawater, and this statistical significance is attributable to the correlation between HAdV and SC (Table 1). Table 2 reports the calculation of HAdV to indicator ratio, using different indicator data (EC, IE, SC) in each water type. Concerning river water, only the data on River No 1 were sufficient for a goodness-of-fit analysis. The ratios are point-estimated as average and IC95% from the
Fig. 1. Concentrations of biological agents relating to the different matrices (geometric mean and standard deviation). HAdV stands for concentration of human adenovirus, EC of Escherichia coli, IE of intestinal enterococci, SC of somatic coliphage, and WWTP for wastewaters treatment plant.
1472
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
Table 1 Multiple regression among HAdV and indicators quantitative data for the different matrices.
WWTP entry WWTP exit River no 1 River no 2 Seawater
HAdV (Log CG/l ± SD)
EC Regression (Log CFU/100 ml ± SD) (HAdV-EC)
IE (Log CFU/100 ml ± SD)
Regression (HAdV-IE)
SC (Log PFU/100 ml ± SD)
Regression (HAdV-SC)
Multiple regression
8.79 ± 1.19 6.70 ± 1.77 5.66 ± 1.59
6.47 ± 0.75 4.77 ± 0.75 1.98 ± 0.90
p = 0.793 p = 0.426 p = 0.709
5.85 ± 0.69 4.26 ± 0.68 1.64 ± 0.40
p = 0.413 p = 0.058 p = 0.489
6.19 ± 0.40 4.01 ± 0.53 1.32 ± 0.57
p = 0.738 p = 0.721 p = 0.922
p = 0.498 p = 0.078 p = 0.907
3.33 ± 1.00
2.09 ± 0.33
p = 0.205
1.94 ± 0.79
p = 0.331
2.24 ± 0.79
p = 0.006
p = 0.035
2.56 ± 0.62
0.26 ± 0.46
p = 0.190
0.09 ± 0.32
p = 0.778
0.21 ± 0.59
p = 0.002
p = 0.023
In bold the value of p with significant relationship (p b 0.05).
collected data, and modelled as a best-fit probability distribution function using EasyFit. From the point-estimate evaluation, on average we observed an increase in the ratios between HAdV and all three indicators from the WWTP entry (6.23 × 102, 1.88 × 103 and 1.15 × 103 for EC, IE and SC, respectively) to the WWTP exit (7.23 × 103, 6.96 × 103 and 6.21 × 103 for EC, IE and SC, respectively) as far as the river waters (1.66 × 104, 1.68 × 104 and 3.99 × 104 for EC, IE and SC, respectively). For seawaters, the average ratio values were included between the WWTP entry and exit values. Fig. 2 represents the data of the HAdV to indicators ratios by boxplot, in order to illustrate the variability of each ratio in terms of the interquartile range. These ratios vary considerably, with on average a 102 order of magnitude of difference. Ratio value modelling was developed to consider the variability and uncertainty of point-estimate values. The best fitted models simulated the observed ratio with reasonable accuracy in all the tested water types (KS test, p N 0.05), as shown in the fourth column in Table 2. In the WWTP entrance, WWTP exit and
river waters, the ratio fitted with lognormal (3P) distribution and Weibull (3P) distributions. In the seawaters, the best-fitting distributions were Weibull (a continuous probability distribution), Frechet (an inverse Weibull distribution) and Gamma (a two-parameter family of continuous probability distribution) for HAdV:IE, HAdV:EC, and HAdV:SC, respectively (Table 2). 4. Discussion Our data are in accordance with other studies on the presence and concentrations of HAdV in water environments, in terms of the high positivity of samples both at the entrance (100%) and exit (94%) of WWTP (Fong et al., 2010; Myrmel et al., 2015; Iaconelli et al., 2017). However, our quantitative molecular data were slightly higher, with a difference of about 1 Log, especially for the entry samples which is probably due to different sources of contamination. Our data for the two rivers and seawater samples are in accordance with other published monitoring. The frequency of positive river samples that we
Table 2 Calculation of HAdV to indicator ratios using different indicators data (EC, IE, SC) in each water type, considering both point estimates (second column) and best-fitted distributions (third column) with the results of KS test in the fourth column. Number of observations used to calculate the ratios are reported in the second column. Number of observations
Point-estimate ratio (average and IC95%)
Ratio modelled as probability density function (distribution and parameters)
KS goodness-of-fit test
WWTP entrance HAdV:EC ratio
52
HAdV:IE ratio
52
HAdV:SC ratio
52
6,23E + 02 (2,05E-01 - 6,81E + 03) 1,88E + 03 (5,73E-01 - 1,92E + 04) 1,15E + 03 (4,41E-01 - 8,66E + 03)
Lognormal 3P (σ = 4,0003; μ = 2,6541; γ = 0,16495) Lognormal 3P (σ = 3,4162; μ = 4,2599; γ = 0,36871) Lognormal (σ = 2,9076; μ = 3,6656)
0,09209 (p = 0.73518) 0,07299 (p = 0,92558) 0,12711 (p = 0,3411)
WWTP exit HAdV:EC ratio
52
HAdV:IE ratio
52
HAdV:SC ratio
52
7,23E + 03 (1,03E-03 - 3,13E + 04) 6,96E + 03 (1,18E-03 - 1,45E + 04) 6,21E + 03 (5,30E-03 - 4,01E + 04)
Weibull (α = 0,2895; β = 45,756) Weibull (α = 0,30528; β = 136,45) Lognormal (σ = 4,1666; μ = 3,8805)
0,08931 (p = 0,76778) 0,09827 (p = 0,66037) 0,06945 (p = 0,94843)
River HAdV:EC ratio
26
HAdV:IE ratio
26
HAdV:SC ratio
26
1,66E + 04 (5,00E-02 - 1,16E + 05) 1,68E + 04 (9,96E-02 - 1,19E + 05) 3,99E + 04 (7,19E-01 - 2,69E + 05)
Weibull (α = 0,38245; β = 4681,9) Weibull 3P (α = 0,25735; β = 3016,3; γ = 0,05714) Weibull (α = 0,28476; β = 11807)
0,13359 (p = 0.69313) 0,18384 (p = 0,30458) 0,13258 (p = 0,70188)
Seawater HAdV:EC ratio
43
HAdV:IE ratio
27
HAdV:SC ratio
30
2,65E + 03 (1,33E + 00 - 1,85E + 04) 4,17E + 03 (1,33E + 00 - 4,26E + 04) 3,39E + 03 (3,35E-01 - 3,04E + 04)
Frechet (α = 0,57852; β = 5,3133) Weibull (α = 0,43457; β = 170,2) Gamma (α = 0,31765; β = 473,7)
0,1348 (p = 0,39505) 0,14052 (p = 0,63315) 0,14127 (p = 0,60436)
For the log-normal 3P, μ and σ are respectively the mean and the standard deviation of the associated normal (γ is the location parameter for the log normal 3P). For the Weibull, Frechet and gamma distributions, α is the shape parameter and β is the scale parameter (γ is the location parameter for the Weibull 3P).
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
1473
Fig. 2. HAdV: indicator ratios in different environmental matrices (minimum, 1° quartile, median, 3° quartile, maximum). HAdV stands for human adenovirus, EC for Escherichia coli, IE for intestinal enterococci, SC for somatic coliphage, and WWTP for wastewaters treatment plant.
found, 92%–72% respectively, and genome quantity is likely due to the distinct levels of contamination of these surface waters due to the possible presence of anthropic sources, as reported by La Rosa et al. (2017) and Marcheggiani et al. (2015) for the river Tiber, or by other international studies (Haramoto et al., 2007; Albinana-Gimenez et al., 2009). In seawater, the low positivity (21%) and the concentrations of viral DNA, of about 102 GC/l, revealed a dilution of HAdV from the pollution sources up to the seawaters, as widely reported (Calgua et al., 2008; Girones et al., 2010). On the basis of the occurrence data on HAdV in surface waters (rivers and sea), our study revealed that samples with a very low FIB level (based on the EU classification for “excellent” water quality, see Section 2.5 Data Analysis) showed a viral contamination by HAdV. These results confirm the above data on the inadequacy of FIB for viral exposure, especially in waters where the pollution sources are not well identified. This aspect is well-known for bathing waters, where viral pathogens are detected in waters compliant with bacterial indicators (Girones and Bofill-Mas, 2013; Love et al., 2014). Our quantitative data revealed no significant statistical correlation between HAdV and bacterial indicators for any of the water samples, thus confirming their inadequacy to indicate viral contamination. This result is in agreement with other monitoring studies carried out in inland waters (Lee and Kim, 2002; Skraber et al., 2004; Jurzik et al., 2010). These results confirm the suggestions by Silva et al. (2011) concerning the inadequacy of FIB to ensure the virological quality of water. However, in our study HAdV GC correlated significantly with the SC concentration, but only for River No 2 and the seawater samples. This could be explained by the different level of fecal pollution of the analyzed environmental matrices. In fact, River No 2 and the seawater showed lower microbiological contamination than the other water types. Some authors have found a lack of association between SC and
viral pathogen in polluted waters, which it has been attributable to the SC infection ability of different bacterial families, not only the Enterobacteriaceae family (Jurzik et al., 2010). HAdV quantitative data were also used to calculate the ratios with each fecal indicator in each water type, since the pathogen to indicator ratio is not constant along hydrological systems. This is due to the different inherent biological features of pathogens, mainly viruses, which exhibit a lower die-off in natural environments compared to indicators (Silva et al., 2011). These ratios may be valuable for the development of QMRA models, which are based on the concentration of pathogens in the environmental matrix, human exposure to these pathogens, and the dose-response relations specific for each pathogen. In these models, the pathogen concentrations can be estimated through indicators, using conversion ratios available in the literature. This approach is used in the current WHO guidelines (WHO, 2016) and has been adopted in some QMRA studies to estimate the health risk from exposure to drinking waters, considering the risk attributable to water supplies (Howard et al., 2006; Zhou et al., 2014), recreational waters (Eregno et al., 2016; Fuhrimann et al., 2016), wastewater reuse (Silverman et al., 2013; Owusu-Ansah et al., 2017), and the land application of biosolids (Gerba et al., 2008; Tanner et al., 2008). Although the adoption of a pathogen to indicator ratio represents an element of uncertainty in the QMRA model input, it is considered appropriate in order to infer human pathogen concentrations when the fecal contamination of the water sources is predominantly of a human origin (O'Toole et al., 2014). The ratios calculated in our study are reliable in river and seawaters, because of the correlations found between HAdV and SC in waters with a low level of contamination. This is an interesting result, because the new water-related regulations (U.S. EPA, 2018; EU, 2018) include SC as routine indicators. In the future the number of SC data in seawater will therefore increase, as well as the data on EC and IE, thus the
1474
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
availability of the ratio between HAdV and SC will facilitate QMRA applications. For the ratios regarding more polluted waters, the lack of a significant correlation reduces the reliability of their use in QMRA, although probability density functions enable the variability of the data to be included. To the best of our knowledge, QMRA studies using pathogen to indicator ratios tend to focus on norovirus and rotavirus. Only one paper has considered this ratio for HAdV (Silverman et al., 2013), in which the authors applied a QMRA model to assess the health risk from exposure to HAdV during wastewater reuse for irrigation purposes in Ghana, and calculated an average HAdV to EC ratio of 10−3.2 based on their samples. This finding is not in line with our results, as we obtained an average HAdV to EC ratio of 7.23 × 103 at the exit of WWTP. These discordant results may depend on differences in the analytical methods, as well as on differences in the epidemiological context. HAdV levels ranged from 2.80 × 102 to 6.50 × 104 GC/100 ml in wastewaters used for irrigational purposes in Ghana, compared to 1.31 × 103 and 8 × 108 GC/100 ml at the exit of WWTP in the present work. The high HAdV concentrations detected in Italy are in accordance with several monitoring surveys documenting the abundance of HAdV in wastewaters in the European Union (La Rosa et al., 2010; Carducci and Verani, 2013; Myrmel et al., 2015). In this paper, HAdV was studied as a potential indicator of viral pollution and index pathogens for QMRA purposes, but however various limitations on the use of HAdV need highlighting. The limitations of HAdV as an indicator of viral pollution may be related to the analytical methods used for HAdV detection in water. In our study, data were generated following standard analyses, described in the international literature and validated in our laboratory experience (two-step tangential flow, decontamination by chloroform, nucleic extraction using a commercial kit and DNA detection by Hernroth et al., 2002). However, different laboratories may follow different protocols, which could affect the results in terms of quantitative data. The main limitation of HAdV as an index pathogen for QMRA purposes is the fact that HAdV detection in environmental samples is mainly based on a molecular technique (PCR), while HAdV is detected using culture methods (e.g., TCID50: infectious dose for 50% of tissue culture wells) in clinical trials for the development of dose-response curves used in QMRA (Teunis et al., 2016). In addition, HAdV to indicators ratios need to be corrected by a conversion factor (CF), representing the ratio between infectious particles and non-infectious GC. This is generally reported in QMRA studies as 1:700 (McBride et al., 2013; Kundu et al., 2013; Vergara et al., 2016). This harmonization factor is assumed by Bambic et al. (2011) based on a study on primary treated sewage. The authors found a genome to PFU ratio of 1:1000, considering a correspondence of 1 TCID50 to 0.7 PFU. This estimate is clearly imprecise because the CF could be affected by the type of water as well as by the HAdV serotypes, which could enhance the uncertainties in health risk estimation by QMRA. Further investigations are necessary to precisely assess the fraction of detected adenovirus GC that are infective in different environmental matrices. 5. Conclusions HAdV could be a possible indicator for virus contamination in waters and could be used as an index pathogen for QMRA studies. We calculated HAdVs to indicators ratios as point estimates and as probability density functions in order to consider the variability and uncertainty of the data. Further studies will focus on verifying these ratios in larger data sets. The estimated ratios refer to HAdV genomic copies that do not correspond to infective particles. Further specific studies will be carried out, in order to clarify the relationship between biomolecular and infectivity viral data, thus filling the current gap in the scientific literature.
Funding The analytical data derived from several studies, funded by national and international institutions: Tuscany Region (Agreement for the protection of coastal-marine waters, Regional Decree No 14 of 7 October 2014) and EU (VIROBATHE-EU FP6 513648). Acknowledgement We wish to thank to English for Academics (e4ac.com) for editing and proofreading the manuscript. References Albinana-Gimenez, N., Miagostovich, M.P., Calgua, B., Huguet, J.M., Matia, L., Girones, R., 2009. Analysis of adenoviruses and polyomaviruses quantified by qPCR as indicators of water quality in source and drinking water treatment plants. Water Res. 47 (3), 2011–2019. Bambic, D., McBride, G., Miller, W., Wuertz, S., 2011. Quantification of Pathogens and Sources of Microbial Indicators for QMRA in Recreational Waters. Pathogen and Human Health. Final Report. WERF, Water Environment Research Foundation Innovation (PATH2R08 (table 6-1a)). Bofill-Mas, S., Albinana-Gimenez, N., Clemente-Casares, P., Hundesa, A., RodriguezManzano, J., Allard, A., Calvo, M., Girones, R., 2006. Quantification and stability of human adenoviruses and polyomavirus JCPyV in wastewater matrices. Appl. Environ. Microbiol. 72, 7894–7896. Calgua, B., Mengewein, A., Grunert, A., Bofill-Mas, S., Clemente-Casares, P., Hundesa, A., Wyn-Jones, A.P., López-Pila, J.M., Girones, R., 2008. Development and application of a one-step low cost procedure to concentrate viruses from seawater samples. J. Virol. Methods 153, 79–83. Carducci, A., Verani, M., 2013. Effects of bacterial, chemical, physical and meteorological variables on virus removal by a Wastewater Treatment Plant. Food Environ. Virol. 5 (1), 69–76. Carducci, A., Casini, B., Bani, A., Rovini, E., Verani, M., Mazzoni, F., Giuntini, A., 2003. Virological control of groundwater quality using biomolecular tests. Water Sci. Technol. 47 (3), 261–266. Carducci, A., Morici, P., Pizzi, F., Battistini, R., Rovini, E., Verani, M., 2008. Study of the viral removal efficiency in a urban wastewater treatment plant. Water Sci. Technol. 58 (4), 893–897. Carducci, A., Donzelli, G., Cioni, L., Federigi, I., Lombardi, R., Verani, M., 2018. Quantitative microbial risk assessment for workers exposed to bioaerosol in wastewater treatment plants aimed at the choice and setup of safety measures. Int. J. Environ. Res. Public Health 15 (7), 1490. Eregno, F.E., Tryland, I., Tjomsland, T., Myrmel, M., Robertson, L., Heistad, A., 2016. Quantitative microbial risk assessment combined with hydrodynamic modelling to estimate the public health risk associated with bathing after rainfall events. Sci. Total Environ. 548-549, 270–279. EU, 2006. Directive 2006/7/EC of the European Parliament and of the Council of 15 February 2006 Concerning the Management of Bathing Water Quality and Repealing Directive 76/160/EEC. EU, 2018. Proposal for a Directive of the European Parliament and of the Council on the Quality of Water Intended for Human Consumption (Recast). European Commission (COM(2017) 753 final 2017/0332 (COD)). Fong, T.T., Phanikumar, M.S., Xagoraraki, I., Rose, J.B., 2010. Quantitative detection of human adenoviruses in wastewater and combined sewer overflows influencing a Michigan River. Appl. Environ. Microbiol. 76 (3), 715–723. Fongaro, G., Nascimento, M.A., Rigotto, C., Ritterbusch, G., da Silva, A.D., Esteves, P.A., Barardi, C.R., 2013. Evaluation and molecular characterization of human adenovirus in drinking water supplies: viral integrity and viability assays. Virol. J. 28 (10), 166. Fuhrimann, S., Winkler, M.S., Stalder, M., Nguyen-Viet, H., Babu, M., Kabatereine, N.B., Halage, A.A., Utzinger, J., Cissé, G., Nauta, M., 2016. Disease burden due to gastrointestinal pathogens in a wastewater system in Kampala, Uganda. Microb. Risk Anal. 4, 16–28. Gerba, C.P., Castro-del Campo, N., Brooks, J.P., Pepper, I.L., 2008. Exposure and risk assessment of Salmonella in recycled residuals. Water Sci. Technol. 57, 1061–1065. Girones, R., Bofill-Mas, S., 2013. Virus indicators for food and water. In: Cook, N. (Ed.), Viruses in Food and Water: Risks, Surveillance, and Control. Woodhead Publishing Ltd, Philadelphia, pp. 483–509. Girones, R., Ferrus, M.A., Alonso, J.L., Rodriguez-Manzano, J., Calgua, B., Corrêa Ade, A., Hundesa, A., Carratala, A., Bofill-Mas, S., 2010. Molecular detection of pathogens in water—the pros and cons of molecular techniques. Water Res. 44 (15), 4325–4339. Hamza, I.A., Jurzik, L., Stang, A., Sure, K., Uberla, K., Wilhelm, M., 2009. Detection of human viruses in rivers of a densly-populated area in Germany using a virus adsorption elution method optimized for PCR analyses. Water Res. 43 (10), 2657–2668. Haramoto, E., Katayama, H., Oguma, K., Ohgaki, S., 2007. Quantitative analysis of human enteric adenoviruses in aquatic environments. J. Appl. Microbiol. 103, 2153–2159. Hernroth, B.E., Conden-Hansson, A.C., Rehnstam-Holm, A.S., Girones, R., Allard, A.K., 2002. Environmental factors influencing human viral pathogens and their potential indicator organisms in the blue mussel, Mytilus edulis: the first Scandinavian report. Appl. Environ. Microbiol. 68 (9), 4523–4533. Hewitt, J., Greening, G.E., Leonard, M., Lewis, G.D., 2013. Evaluation of human adenovirus and human polyomavirus as indicators of human sewage contamination in the aquatic environment. Water Res. 47, 6750–6761.
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475 Howard, G., Pedley, S., Tibatemwa, S., 2006. Quantitative microbial risk assessment to estimate health risks attributable to water supply: can the technique be applied in developing countries with limited data? J. Water Health 4 (1), 49–65. Iaconelli, M., Muscillo, M., Della Libera, S., Fratini, M., Meucci, L., De Ceglia, M., Giacosa, D., La Rosa, G., 2017. One-year surveillance of human enteric viruses in raw and treated wastewaters, downstream river waters, and drinking waters. Food Environ. Virol. 9 (1), 79–88. https://doi.org/10.1007/s12560-016-9263-3. Epub 2016 Sep 28. Jurzik, L., Hamza, I.A., Puchert, W., Uberla, K., Wilhelm, M., 2010. Chemical and microbiological parameters as possible indicators for human enteric viruses in surface water. Int. J. Hyg. Environ. Health 213 (3), 210–216. Kundu, A., McBride, G., Wuertz, S., 2013. Adenovirus-associated health risks for recreational activities in a multi-use coastal watershed based on site-specific quantitative microbial risk assessment. Water Res. 47 (16), 6309–6325. La Rosa, G., Muscillo, M., 2013. Molecular detection of viruses in water and sewage. In: Cook, Nigel (Ed.), Virus in Food and Water – Risk, Surveillance and Control. Woodhead Publishing. ISBN: 9780857094308. La Rosa, G., Pourshaban, M., Iaconelli, M., Muscillo, M., 2010. Quantitative real-time PCR of enteric viruses in influent and effluent samples from wastewater treatment plants in Italy. Ann. Ist. Super. Sanita 46 (3), 266–273. La Rosa, G., Sanseverino, I., Della Libera, S., Iaconelli, M., Ferrero, V.E.V., Barra Caracciolo, A., Lettieri, T., 2017. The impact of anthropogenic pressure on the virological quality of water from the Tiber River, Italy. Lett. Appl. Microbiol. 65 (4), 298–305. Lee, S.H., Kim, S.J., 2002. Detection of infectious enteroviruses and adenoviruses in tap water in urban areas in Korea. Water Res. 36, 248–256. Lin, J., Ganesh, A., 2013. Water quality indicators: bacteria, coliphages, enteric viruses. Int. J. Environ. Health Res. 23 (6), 484–506. Linden, K.G., Thurston, J., Schaefer, R., Malley, J.P., 2007. Enhanced UV inactivation of adenoviruses under polychromatic UV lamps. Appl. Environ. Microbiol. 73, 7571–7574. Love, D.C., Rodriguez, R.A., Gibbons, C.D., Griffith, J.F., Yu, Q., Stewart, J.R., Sobsey, M.D., 2014. Human viruses and viral indicators in marine water at two recreational beaches in Southern California, USA. J. Water Health 12 (1), 136–150. Marcheggiani, S., D'Ugo, E., Puccinelli, C., Giuseppetti, R., D'Angelo, A.M., Gualerzi, C.O., Spurio, R., Medlin, L.K., Guillebault, D., Baudart-Lenfant, J., Weigel, W., Helmi, K., Mancini, L., 2015. Detection of emerging and re-emerging pathogens in surface waters close to an urban area. Int. J. Environ. Res. Public Health 12 (5), 5505–5527. McBride, G.B., Stott, R., Miller, W., Bambic, D., Wuertz, S., 2013. Discharge-based QMRA for estimation of public health risks from exposure to stormwater-borne pathogens in recreational waters in the United States. Water Res. 47 (14), 5282–5297. Myrmel, M., Lange, H., Rimstad, E., 2015. A 1-year quantitative survey of noro-, adeno-, human boca-, and hepatitis E viruses in raw and secondarily treated sewage from two plants in Norway. Food Environ. Virol. 7, 213–223. Nordgren, J., Matussek, A., Mattssond, A., Svenssone, L., Lindgren, P.E., 2009. Prevalence of norovirus and factors influencing virus concentrations during one year in a full-scale wastewater treatment plant. Water Res. 43, 1117–1125. O'Toole, J., Sinclair, M., Barker, F., Leder, K., 2014. Advice to risk assessors modeling viral health risk associated with household graywater. Risk Anal. 34 (5). https://doi.org/ 10.1111/risa.12142. Owusu-Ansah, E.D.J., Sampson, A., Amponsah, S.K., Abaidoo, R.C., Dalsgaard, A., Hald, T., 2017. Probabilistic quantitative microbial risk assessment model of norovirus from wastewater irrigated vegetables in Ghana using genome copies and fecal indicator ratio conversion for estimating exposure dose. Sci. Total Environ. 601–602, 1712–1719. Poma, H.R., Gutiérrez Cacciabue, D., Garcé, B., Gonzo, E.E., Rajal, V.B., 2012. Towards a rational strategy for monitoring of microbiological quality of ambient waters. Sci. Total Environ. 433, 98–109.
1475
Silva, H.D., García-Zapata, M.T.A., Anunciação, C.E., 2011. Why the use of adenoviruses as water quality virologic marker? Food Environ. Virol. 3, 138–140. Silverman, A.I., Akrong, M.O., Amoah, P., Drechsel, P., Nelson, K.L., 2013. Quantification of human norovirus GII, human adenovirus, and fecal indicator organisms in wastewater used for irrigation in Accra, Ghana. J. Water Health 11 (3), 473–488. Sinclair, R.G., Jones, E.L., Gerba, C.P., 2009. Viruses in recreational water-borne disease outbreaks: a review. J. Appl. Microbiol. 107 (6), 1769–1780. Skraber, S., Gassilloud, B., Gantzer, C., 2004. Comparison of coliforms and coliphages as tools for assessment of viral contamination in river water. Appl. Environ. Microbiol. 70, 3644–3649. Tanner, B.D., Brooks, J.P., Gerba, C.P., Haas, C.N., Josephson, K.L., Pepper, I.L., 2008. Estimated occupational risk from bioaerosols generated during land application of Class B biosolids. J. Environ. Qual. 37, 2311–2321. Teunis, P., Schijven, J., Rutjes, S., 2016. A generalized dose-response relationship for adenovirus infection and illness by exposure pathway. Epidemiol. Infect. 30, 1–13. Thurston-Enriquez, J.A., Haas, C.N., Jacangelo, J., Gerba, C.P., 2005. Inactivation of enteric adenovirus and feline calicivirus by chlorine dioxide. Appl. Environ. Microbiol. 71, 3100–3105. U.S. EPA, 2016. Summary of Nominations for the Fourth Contaminant Candidate List (CCL 4). Office of Water (4607M). U. S. Environmental Protection Agency, Office of Water (EPA 815-R-16-006). U.S. EPA, 2018. Five-Year Review of the 2012 Recreational Water Quality Criteria. U. S. Environmental Protection Agency, Office of Water (EPA 823-R-18-001). van Heerden, J., Ehlers, M.M., Grabow, W.O.K., 2005. Detection and risk assessment of adenoviruses in swimming pool water. J. Appl. Microbiol. 99 (5), 1256–1264. Verani, M., Di Giuseppe, G., Tammaro, C., Carducci, A., 2016. Investigating the role of Acanthamoeba polyphaga in protecting human adenovirus from water disinfection treatment. Eur. J. Protistol. 54, 11–18. Verbyla, M.E., Symonds, E.M., Kafle, R.C., Cairns, M.R., Iriarte, M., Mercado Guzmán, A., Coronado, O., Breitbart, M., Ledo, C., Mihelcic, J.R., 2016. Managing microbial risks from indirect wastewater reuse for irrigation in urbanizing watersheds. Environ. Sci. Technol. 50 (13), 6803–6813. Vergara, G.G.R.V., Rose, J.B., Gin, K.Y.H., 2016. Risk assessment of noroviruses and human adenoviruses in recreational surface waters. Water Res. 103, 276–282. Vieira, C.B., de Abreu Corrêa, A., de Jesus, M.S., Luz, S.L., Wyn-Jones, P., Kay, D., Vargha, M., Miagostovich, M.P., 2016. Viruses surveillance under different season scenarios of the Negro River Basin, Amazonia, Brazil. Food Environ. Virol. 8 (1), 57–69. WHO, 2011. Guidelines for Drinking-Water Quality. World Health Organization, Geneva, Switzerland. WHO, 2016. Quantitative Microbial Risk Assessment: Application for Water Safety Management. World Health Organization, Geneva, Switzerland. Wyer, M.D., Wyn-Jones, A.P., Kay, D., Au-Yeung, H.-K.C., Girones, R., Lopez-Pila, J., de Roda Husman, A.M., Rutjes, S., Schneider, O., 2012. Relationships between human adenoviruses and faecal indicator organisms in European recreational waters. Water Res. 46, 4130–4141. Wyn-Jones, A.P., Carducci, A., Cook, N., D'Agostino, M., Divizia, M., Fleischer, J., Gantzer, C., Gawler, A., Girones, R., Höller, C., de Roda Husman, A.M., Kay, D., Kozyra, I., López-Pila, J., Muscillo, M., Nascimento, M.S., Papageorgiou, G., Rutjes, S., Sellwood, J., Szewzyk, R., Wyer, M., 2011. Surveillance of adenoviruses and noroviruses in European recreational waters. Water Res. 45 (3), 1025–1038. Zhou, L., Echigo, S., Ohkouchi, Y., Itoh, S., 2014. Quantitative microbial risk assessment of drinking water treated with advanced water treatment process. J. Water Supply Res Technol. 63 (2), 114–123.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Human adenoviruses as waterborne index pathogens and their use for Quantitative Microbial Risk Assessment Marco Verani a, Ileana Federigi a,⁎, Gabriele Donzelli a, Lorenzo Cioni b, Annalaura Carducci a a b
Laboratory of Hygiene and Environmental Virology, Department of Biology, University of Pisa, Via S. Zeno 37, 56127 Pisa, Italy Scuola Normale Superiore, P.zza dei Cavalieri, 7, 56126 Pisa, Italy
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• HAdV quantification in different water samples in parallel with fecal indicators • 64% of samples positive for HAdV, decreasing from WWTP (100%) to seawaters (21%) • HAdV occurrence in seawater samples complied with excellent EU bathing water criteria • Significant correlation between HAdV and somatic coliphages in less polluted waters • HAdV to indicators ratios modelled as probability density functions for QMRA
a r t i c l e
i n f o
Article history: Received 31 May 2018 Received in revised form 22 September 2018 Accepted 22 September 2018 Available online 25 September 2018 Editor: Paola Verlicchi Keywords: Human adenovirus Escherichia coli Intestinal enterococci Somatic coliphages QMRA Water safety
a b s t r a c t The current microbial water quality standards are based on the monitoring of fecal indicator organisms, which are mainly bacterial indicators (i.e., Escherichia coli, intestinal enterococci), however epidemiological data indicate that viruses are important etiological agents of waterborne illnesses. Among waterborne viruses, human adenovirus can be considered as an index pathogen, owing to its abundance in sewage and persistence in the environment, as well as its potential infectivity. In this study, data on human adenoviruses from different water matrices (the entrance and exit of a water treatment plant, rivers and seawaters) were analyzed, in parallel with traditional fecal bacterial indicators and somatic coliphages. The results showed a 64% frequency of positive adenovirus samples, decreasing from the sewage system (100% at the entrance and 94% at the exit) to rivers (92% and 72% for different rivers) and seawater (21%). Adenovirus concentrations showed a significant correlation with somatic coliphages in one river and seawater, thus supporting the recent inclusion of coliphages as viral indicators in water safety guidelines. The data collected were used to estimate adenovirus to indicator ratios, which could be used as input in Quantitative Microbial Risk Assessment (QMRA) studies. © 2018 Published by Elsevier B.V.
1. Introduction ⁎ Corresponding author. E-mail addresses: [email protected] (M. Verani), [email protected] (I. Federigi), [email protected] (G. Donzelli), [email protected] (L. Cioni), [email protected] (A. Carducci).
https://doi.org/10.1016/j.scitotenv.2018.09.295 0048-9697/© 2018 Published by Elsevier B.V.
The fecal pollution of waters used for different purposes (potable, reuse or recreation) is a public health issue. The consumption of, or contact with, contaminated waters is associated with outbreaks caused by
1470
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
waterborne pathogens, above all enteric viruses and protozoan parasites (Love et al., 2014). Of the enteric viruses, special attention has been given to the human adenovirus (HAdV) owing to its diffusion, epidemiological importance and easy detection. HAdV is a member of the genus Mastadenovirus in the Adenoviridae family, which comprises at least 51 serotypes, which are classified into seven species (A–G). HAdV is responsible for a wide range of health problems (including respiratory, gastrointestinal, and urinary infections), but also for a large number of asymptomatic infections, and all HAdV types can be excreted in high concentration into feces from infected individuals (Silva et al., 2011). These features have justified the inclusion of HAdVs in the US Environmental Protection Agency's water contaminant candidate list (U.S. EPA, 2016). Moreover, many studies have also suggested using HAdV as an indicator of the viral contamination in water (Wyer et al., 2012; Hewitt et al., 2013), or of the efficiency of viral removal in wastewater depuration plants (Carducci et al., 2008; La Rosa et al., 2010). These recommendations regarding HAdVs are made in relation to several specific qualities: - their high frequency retrieval and persistence in different water matrices such as wastewater effluents, rivers, seawaters and drinking waters (Haramoto et al., 2007; Wyn-Jones et al., 2011; Fongaro et al., 2013); - their resistance to water treatment/disinfection highlighted by studies that have revealed the presence of viable particles in waters treated by chlorine (van Heerden et al., 2005; Thurston-Enriquez et al., 2005) and their resistance to ultraviolet (UV) disinfection (Linden et al., 2007); - their higher frequency and abundance in waters compared to other enteric viruses, with genomic copies (GC) concentrations ranging from ~108 GC/100 ml in urban sewage (La Rosa et al., 2010; Carducci and Verani, 2013) to ~103 GC/100 ml in surface waters (Poma et al., 2012; Vieira et al., 2016); - their association with disease outbreaks in recreational waters (Sinclair et al., 2009); - the stability of their genome in comparison to RNA viruses (i.e. norovirus) guaranteeing a more “conservative” approach for risk assessment; - the possibility of cultivation (for most of the strains) and of biomolecular detection which enables the proportion of infective particles on genome units to be estimated. HAdV is thus considered as a promising indicator of viral pollution in different water-related area (Albinana-Gimenez et al., 2009; Hewitt et al., 2013) and is used as an index pathogen in Quantitative Microbial Risk Assessment (QMRA) in many different settings such as natural and treated recreational waters (Kundu et al., 2013), reused waters (Verbyla et al., 2016) and bioaerosol from wastewater treatment plants (Carducci et al., 2018). Nevertheless, current microbial water quality standards are mainly based on the monitoring of fecal indicator bacteria because of the low complexity and inexpensive detection methods. These microorganisms represent indicators of fecal contamination from warm-blooded animals (total and fecal coliform, E. coli, intestinal enterococci, Clostridium perfringens) some of which are specific indicators of human fecal pollution (E. coli and intestinal enterococci) (WHO, 2011). In addition, alternative microorganisms have been proposed as indicators used to assess viral water safety, such as somatic coliphage, which were added to the list of indicators in recent guidelines for recreational waters (U.S. EPA, 2018) and in the proposed EU regulation for drinking waters safety (EU, 2018). In the context of public health protection, these parameters are still an issue of great controversy, because of the lack of consistent correlation between bacterial indicators and enteric viral pathogens, such as the Adenoviridae family (Girones and Bofill-Mas, 2013; Love et al., 2014) and the reliability of somatic coliphage used as viral indicators
which is still under discussion (Lin and Ganesh, 2013). Considering the possible use of HAdV as both an indicator of viral pollution and index pathogen for QMRA, in this study we evaluated its relationship to other indicators. We thus analyzed the results of the study of different types of water (sewage, sea and rivers) conducted by the Hygiene and Environmental Virology Laboratory of the University of Pisa. The aim was to compare the occurrence and the amount of HAdV with those of fecal bacteria indicators - FIB (E. coli - EC, and Intestinal Enterococci IE) and viral indicators (Somatic Coliphages - SC). We also used the collected data to determine the ratios (point estimates and modelled as probability density functions) between HAdV and EC, IE, and SC, to be used for QMRA purposes, in order to estimate the HAdV concentrations on the basis of the indicator's concentrations. This approach was included by WHO (2016) in the current guidelines for the QMRA application to Water Safety Management. 2. Material and methods 2.1. Samples During several environmental monitoring campaigns, that were part of the daily sampling activities of our laboratory, we analyzed 241 water samples that were collected in central-northern Italy. All samples were collected in sterile containers, transported to the laboratory in cooler boxes containing ice and analyzed within 24 h. In particular: 1. 104 samples (52 of which were from the plant entry and 52 from the plant exit) were taken from a municipal wastewater treatment plant (WWTP), situated in Pisa; 2. 100 seawater samples were derived from monitoring campaigns on the Tyrrhenian coast, from bathing waters located at river mouths; 3. 37 samples were taken from 2 rivers of which 26 were from a small river that runs through an alluvial plain in northern Tuscany impacted by urban sewage and diffuse pollution sources (River No 1), and 11 were taken from the river Pescia, a tributary of the river Arno (River No 2) affected above all by agricultural runoff. For HAdV detection, 10 L of each water sample were analyzed except for the wastewater treatment effluent, for which the sampling volume was 1 L. For the determination of EC, IE, and SC, a sample volume of 100 ml was analyzed. 2.2. Samples treatment for HAdV detection Samples were concentrated to 40 ml volume, using a two-step tangential-flow ultrafiltration already previously described, with a recovery efficiency up to 100% (Carducci et al., 2003; La Rosa and Muscillo, 2013). Briefly, the samples were filtered in two steps using two devices: an apparatus equipped with polysulfone membrane with a 10,000 MW exclusion size to give a final 250 ml sample eluted in 3% beef extract at pH 9, then the second concentration was performed with a Mini-ultrasette apparatus, to obtain the concentrated sample. After a pH neutralization, the concentrated samples were treated with chloroform to eliminate bacteria and enveloped viruses (Carducci et al., 2008). 2.3. Biomolecular analysis Viral DNA was extracted from concentrated samples using a QIAamp DNA kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The extracted nucleic acid was assayed by real time quantitative PCR (qPCR) (Bofill-Mas et al., 2006). All reactions were performed in triplicate in 96-well optical plates using an ABI 7300 sequence detector system (Applied Biosystems, Foster City, California) and, for quantification, viral GC numbers were obtained from the standard curve dilution
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
series equation (range = 102–109). The standard curve was constructed by cloning the entire hexon region of Ad41 into pBR322 (Verani et al., 2016). In each plate, a positive control and a negative control were tested separately. A control for the presence of enzymatic inhibitors was also performed by analysing both undiluted and 10-fold diluted extracted DNA for each sample (La Rosa and Muscillo, 2013). To determine if the negative signals in the diluted extracts were due to low DNA amount, these DNA extracts were spiked with known concentration of plasmid DNA (103 copies/reaction) (internal control). An increase in the threshold cycles (Ct) of spiked DNA extracts was considered to indicate qPCR inhibition (Hamza et al., 2009). The detection limit for quantitative PCR was ~1 GC/reaction. 2.4. Fecal indicators Fecal indicator bacteria were analyzed according to ISO 9308-3 (1998) for EC, and ISO 7899-1 (1999) for IE. Cell density was enumerated using a microplate method, and counted as the most probable number (MPN)/100 ml of sample. SC titres were determined following the ISO 10705-2 (2001) cultural method based on double agar layer plaque assay using E. coli as host strain (E. coli strain C (ATCC 13706) for river and seawaters and E. coli strain CN (ATCC 700078) for sewage samples). The results were counted as the Plaque Forming Unit (PFU)/ 100 ml of sample. 2.5. Data analysis The microbial concentrations data were Log10 transformed and the geometrical means were calculated. MedCalc Statistical Software 14 (MedCalc Software, Ostend, Belgium) was used for multiple regression analyses in order to assess the potential correlations between the HAdV concentrations and the microbial indicators (EC, IE, and SC) in each type of water. For the regression analyses, a ‘viral load’ corresponding to half of the detection limit was attributed to samples with a negative qPCR (Nordgren et al., 2009). Boxplot graphs were generated with Excel for Windows (Microsoft Office Excel 2016, Redmond, Washington, USA). Each graph represents the whiskers of the minimum and maximum values and the boxes from the 1° and 3° quartile of the data. To compare HAdV occurrence in waters with different levels of bacterial pollution, river and seawater samples were divided based on the threshold for “excellent” water quality according to the EU bathing water directive, with bacterial concentrations expressed in Colony Forming Units (CFU)/100 ml (EU, 2006): 100 CFU/100 ml for IE and 250 CFU/100 ml for EC in river waters, and 200 CFU/100 ml for IE and 500 CFU/100 ml for EC in seawater. A chi square test was then
1471
performed using MedCalc software to compare the frequency of HAdV positivity in waters above the bacterial threshold with the ones below. 2.6. Estimation of ratios between HAdV and fecal indicators Only samples with quantifiable concentrations of at least one parameter were considered for the ratios calculation. The ratios between GC of HAdV and indicators in each type of water were estimated through a best-fit probability distribution function, selected using EasyFit software (MathWave Technologies, accessed at http:// www.mathwave.com/en/home.html on May 2018) according to the Kolmogorov-Smirnov (KS) test as a goodness of fit statistical test. 3. Results The overall frequency of HAdV positive samples was 64%, with a significant decreasing trend from untreated sewages to the seawater: 100% in WWTP entry, 94% in WWTP exit, 92% in River No 1, 72% in River No 2 and 21% in seawater (p b 0.001). A similar trend was found among the geometric means of the concentrations: 8.78 ± 1.19 Log GC/l for WWTP entry, 6.70 ± 1.77 Log GC/l for WWTP exit, 5.66 ± 1.59 Log GC/l for River No 1, 3.33 ± 1.00 Log GC/l for River No 2 and 2.56 ± 0.62 for seawater. These different concentrations are in accordance with the progressive dilution effect of the fecal contamination from sewages to the cleaner surface waters. The control for the presence of enzymatic inhibitors showed no inhibitory effects for DNA amplification in qPCR. For the FIB and SC concentrations, the observed trend among rivers was different, because the microbial pollution in River No 2 was higher than in River No 1 (Fig. 1). In seawater, all samples were below the threshold for “excellent” water quality (EU, 2006) based on FIB (see Section 2.5 Data Analysis), however they were positive for HAdV (21%). In River No 1 and River No 2, 77.8% (17/22) and 75% (6/8) of samples, respectively, were positive for adenovirus, and they exhibited FIB contamination under the threshold. Both in rivers and seawaters, the HAdV frequency was not correlated to water quality based on FIB (chi square, p = 0.92 for River No1 and p = 0.20 for River No2, p = 0.32 for seawater). The results of multiple regression analysis showed a significant relationship among HAdV and indicators (p b 0.05) in River No 2 and seawater, and this statistical significance is attributable to the correlation between HAdV and SC (Table 1). Table 2 reports the calculation of HAdV to indicator ratio, using different indicator data (EC, IE, SC) in each water type. Concerning river water, only the data on River No 1 were sufficient for a goodness-of-fit analysis. The ratios are point-estimated as average and IC95% from the
Fig. 1. Concentrations of biological agents relating to the different matrices (geometric mean and standard deviation). HAdV stands for concentration of human adenovirus, EC of Escherichia coli, IE of intestinal enterococci, SC of somatic coliphage, and WWTP for wastewaters treatment plant.
1472
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
Table 1 Multiple regression among HAdV and indicators quantitative data for the different matrices.
WWTP entry WWTP exit River no 1 River no 2 Seawater
HAdV (Log CG/l ± SD)
EC Regression (Log CFU/100 ml ± SD) (HAdV-EC)
IE (Log CFU/100 ml ± SD)
Regression (HAdV-IE)
SC (Log PFU/100 ml ± SD)
Regression (HAdV-SC)
Multiple regression
8.79 ± 1.19 6.70 ± 1.77 5.66 ± 1.59
6.47 ± 0.75 4.77 ± 0.75 1.98 ± 0.90
p = 0.793 p = 0.426 p = 0.709
5.85 ± 0.69 4.26 ± 0.68 1.64 ± 0.40
p = 0.413 p = 0.058 p = 0.489
6.19 ± 0.40 4.01 ± 0.53 1.32 ± 0.57
p = 0.738 p = 0.721 p = 0.922
p = 0.498 p = 0.078 p = 0.907
3.33 ± 1.00
2.09 ± 0.33
p = 0.205
1.94 ± 0.79
p = 0.331
2.24 ± 0.79
p = 0.006
p = 0.035
2.56 ± 0.62
0.26 ± 0.46
p = 0.190
0.09 ± 0.32
p = 0.778
0.21 ± 0.59
p = 0.002
p = 0.023
In bold the value of p with significant relationship (p b 0.05).
collected data, and modelled as a best-fit probability distribution function using EasyFit. From the point-estimate evaluation, on average we observed an increase in the ratios between HAdV and all three indicators from the WWTP entry (6.23 × 102, 1.88 × 103 and 1.15 × 103 for EC, IE and SC, respectively) to the WWTP exit (7.23 × 103, 6.96 × 103 and 6.21 × 103 for EC, IE and SC, respectively) as far as the river waters (1.66 × 104, 1.68 × 104 and 3.99 × 104 for EC, IE and SC, respectively). For seawaters, the average ratio values were included between the WWTP entry and exit values. Fig. 2 represents the data of the HAdV to indicators ratios by boxplot, in order to illustrate the variability of each ratio in terms of the interquartile range. These ratios vary considerably, with on average a 102 order of magnitude of difference. Ratio value modelling was developed to consider the variability and uncertainty of point-estimate values. The best fitted models simulated the observed ratio with reasonable accuracy in all the tested water types (KS test, p N 0.05), as shown in the fourth column in Table 2. In the WWTP entrance, WWTP exit and
river waters, the ratio fitted with lognormal (3P) distribution and Weibull (3P) distributions. In the seawaters, the best-fitting distributions were Weibull (a continuous probability distribution), Frechet (an inverse Weibull distribution) and Gamma (a two-parameter family of continuous probability distribution) for HAdV:IE, HAdV:EC, and HAdV:SC, respectively (Table 2). 4. Discussion Our data are in accordance with other studies on the presence and concentrations of HAdV in water environments, in terms of the high positivity of samples both at the entrance (100%) and exit (94%) of WWTP (Fong et al., 2010; Myrmel et al., 2015; Iaconelli et al., 2017). However, our quantitative molecular data were slightly higher, with a difference of about 1 Log, especially for the entry samples which is probably due to different sources of contamination. Our data for the two rivers and seawater samples are in accordance with other published monitoring. The frequency of positive river samples that we
Table 2 Calculation of HAdV to indicator ratios using different indicators data (EC, IE, SC) in each water type, considering both point estimates (second column) and best-fitted distributions (third column) with the results of KS test in the fourth column. Number of observations used to calculate the ratios are reported in the second column. Number of observations
Point-estimate ratio (average and IC95%)
Ratio modelled as probability density function (distribution and parameters)
KS goodness-of-fit test
WWTP entrance HAdV:EC ratio
52
HAdV:IE ratio
52
HAdV:SC ratio
52
6,23E + 02 (2,05E-01 - 6,81E + 03) 1,88E + 03 (5,73E-01 - 1,92E + 04) 1,15E + 03 (4,41E-01 - 8,66E + 03)
Lognormal 3P (σ = 4,0003; μ = 2,6541; γ = 0,16495) Lognormal 3P (σ = 3,4162; μ = 4,2599; γ = 0,36871) Lognormal (σ = 2,9076; μ = 3,6656)
0,09209 (p = 0.73518) 0,07299 (p = 0,92558) 0,12711 (p = 0,3411)
WWTP exit HAdV:EC ratio
52
HAdV:IE ratio
52
HAdV:SC ratio
52
7,23E + 03 (1,03E-03 - 3,13E + 04) 6,96E + 03 (1,18E-03 - 1,45E + 04) 6,21E + 03 (5,30E-03 - 4,01E + 04)
Weibull (α = 0,2895; β = 45,756) Weibull (α = 0,30528; β = 136,45) Lognormal (σ = 4,1666; μ = 3,8805)
0,08931 (p = 0,76778) 0,09827 (p = 0,66037) 0,06945 (p = 0,94843)
River HAdV:EC ratio
26
HAdV:IE ratio
26
HAdV:SC ratio
26
1,66E + 04 (5,00E-02 - 1,16E + 05) 1,68E + 04 (9,96E-02 - 1,19E + 05) 3,99E + 04 (7,19E-01 - 2,69E + 05)
Weibull (α = 0,38245; β = 4681,9) Weibull 3P (α = 0,25735; β = 3016,3; γ = 0,05714) Weibull (α = 0,28476; β = 11807)
0,13359 (p = 0.69313) 0,18384 (p = 0,30458) 0,13258 (p = 0,70188)
Seawater HAdV:EC ratio
43
HAdV:IE ratio
27
HAdV:SC ratio
30
2,65E + 03 (1,33E + 00 - 1,85E + 04) 4,17E + 03 (1,33E + 00 - 4,26E + 04) 3,39E + 03 (3,35E-01 - 3,04E + 04)
Frechet (α = 0,57852; β = 5,3133) Weibull (α = 0,43457; β = 170,2) Gamma (α = 0,31765; β = 473,7)
0,1348 (p = 0,39505) 0,14052 (p = 0,63315) 0,14127 (p = 0,60436)
For the log-normal 3P, μ and σ are respectively the mean and the standard deviation of the associated normal (γ is the location parameter for the log normal 3P). For the Weibull, Frechet and gamma distributions, α is the shape parameter and β is the scale parameter (γ is the location parameter for the Weibull 3P).
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
1473
Fig. 2. HAdV: indicator ratios in different environmental matrices (minimum, 1° quartile, median, 3° quartile, maximum). HAdV stands for human adenovirus, EC for Escherichia coli, IE for intestinal enterococci, SC for somatic coliphage, and WWTP for wastewaters treatment plant.
found, 92%–72% respectively, and genome quantity is likely due to the distinct levels of contamination of these surface waters due to the possible presence of anthropic sources, as reported by La Rosa et al. (2017) and Marcheggiani et al. (2015) for the river Tiber, or by other international studies (Haramoto et al., 2007; Albinana-Gimenez et al., 2009). In seawater, the low positivity (21%) and the concentrations of viral DNA, of about 102 GC/l, revealed a dilution of HAdV from the pollution sources up to the seawaters, as widely reported (Calgua et al., 2008; Girones et al., 2010). On the basis of the occurrence data on HAdV in surface waters (rivers and sea), our study revealed that samples with a very low FIB level (based on the EU classification for “excellent” water quality, see Section 2.5 Data Analysis) showed a viral contamination by HAdV. These results confirm the above data on the inadequacy of FIB for viral exposure, especially in waters where the pollution sources are not well identified. This aspect is well-known for bathing waters, where viral pathogens are detected in waters compliant with bacterial indicators (Girones and Bofill-Mas, 2013; Love et al., 2014). Our quantitative data revealed no significant statistical correlation between HAdV and bacterial indicators for any of the water samples, thus confirming their inadequacy to indicate viral contamination. This result is in agreement with other monitoring studies carried out in inland waters (Lee and Kim, 2002; Skraber et al., 2004; Jurzik et al., 2010). These results confirm the suggestions by Silva et al. (2011) concerning the inadequacy of FIB to ensure the virological quality of water. However, in our study HAdV GC correlated significantly with the SC concentration, but only for River No 2 and the seawater samples. This could be explained by the different level of fecal pollution of the analyzed environmental matrices. In fact, River No 2 and the seawater showed lower microbiological contamination than the other water types. Some authors have found a lack of association between SC and
viral pathogen in polluted waters, which it has been attributable to the SC infection ability of different bacterial families, not only the Enterobacteriaceae family (Jurzik et al., 2010). HAdV quantitative data were also used to calculate the ratios with each fecal indicator in each water type, since the pathogen to indicator ratio is not constant along hydrological systems. This is due to the different inherent biological features of pathogens, mainly viruses, which exhibit a lower die-off in natural environments compared to indicators (Silva et al., 2011). These ratios may be valuable for the development of QMRA models, which are based on the concentration of pathogens in the environmental matrix, human exposure to these pathogens, and the dose-response relations specific for each pathogen. In these models, the pathogen concentrations can be estimated through indicators, using conversion ratios available in the literature. This approach is used in the current WHO guidelines (WHO, 2016) and has been adopted in some QMRA studies to estimate the health risk from exposure to drinking waters, considering the risk attributable to water supplies (Howard et al., 2006; Zhou et al., 2014), recreational waters (Eregno et al., 2016; Fuhrimann et al., 2016), wastewater reuse (Silverman et al., 2013; Owusu-Ansah et al., 2017), and the land application of biosolids (Gerba et al., 2008; Tanner et al., 2008). Although the adoption of a pathogen to indicator ratio represents an element of uncertainty in the QMRA model input, it is considered appropriate in order to infer human pathogen concentrations when the fecal contamination of the water sources is predominantly of a human origin (O'Toole et al., 2014). The ratios calculated in our study are reliable in river and seawaters, because of the correlations found between HAdV and SC in waters with a low level of contamination. This is an interesting result, because the new water-related regulations (U.S. EPA, 2018; EU, 2018) include SC as routine indicators. In the future the number of SC data in seawater will therefore increase, as well as the data on EC and IE, thus the
1474
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475
availability of the ratio between HAdV and SC will facilitate QMRA applications. For the ratios regarding more polluted waters, the lack of a significant correlation reduces the reliability of their use in QMRA, although probability density functions enable the variability of the data to be included. To the best of our knowledge, QMRA studies using pathogen to indicator ratios tend to focus on norovirus and rotavirus. Only one paper has considered this ratio for HAdV (Silverman et al., 2013), in which the authors applied a QMRA model to assess the health risk from exposure to HAdV during wastewater reuse for irrigation purposes in Ghana, and calculated an average HAdV to EC ratio of 10−3.2 based on their samples. This finding is not in line with our results, as we obtained an average HAdV to EC ratio of 7.23 × 103 at the exit of WWTP. These discordant results may depend on differences in the analytical methods, as well as on differences in the epidemiological context. HAdV levels ranged from 2.80 × 102 to 6.50 × 104 GC/100 ml in wastewaters used for irrigational purposes in Ghana, compared to 1.31 × 103 and 8 × 108 GC/100 ml at the exit of WWTP in the present work. The high HAdV concentrations detected in Italy are in accordance with several monitoring surveys documenting the abundance of HAdV in wastewaters in the European Union (La Rosa et al., 2010; Carducci and Verani, 2013; Myrmel et al., 2015). In this paper, HAdV was studied as a potential indicator of viral pollution and index pathogens for QMRA purposes, but however various limitations on the use of HAdV need highlighting. The limitations of HAdV as an indicator of viral pollution may be related to the analytical methods used for HAdV detection in water. In our study, data were generated following standard analyses, described in the international literature and validated in our laboratory experience (two-step tangential flow, decontamination by chloroform, nucleic extraction using a commercial kit and DNA detection by Hernroth et al., 2002). However, different laboratories may follow different protocols, which could affect the results in terms of quantitative data. The main limitation of HAdV as an index pathogen for QMRA purposes is the fact that HAdV detection in environmental samples is mainly based on a molecular technique (PCR), while HAdV is detected using culture methods (e.g., TCID50: infectious dose for 50% of tissue culture wells) in clinical trials for the development of dose-response curves used in QMRA (Teunis et al., 2016). In addition, HAdV to indicators ratios need to be corrected by a conversion factor (CF), representing the ratio between infectious particles and non-infectious GC. This is generally reported in QMRA studies as 1:700 (McBride et al., 2013; Kundu et al., 2013; Vergara et al., 2016). This harmonization factor is assumed by Bambic et al. (2011) based on a study on primary treated sewage. The authors found a genome to PFU ratio of 1:1000, considering a correspondence of 1 TCID50 to 0.7 PFU. This estimate is clearly imprecise because the CF could be affected by the type of water as well as by the HAdV serotypes, which could enhance the uncertainties in health risk estimation by QMRA. Further investigations are necessary to precisely assess the fraction of detected adenovirus GC that are infective in different environmental matrices. 5. Conclusions HAdV could be a possible indicator for virus contamination in waters and could be used as an index pathogen for QMRA studies. We calculated HAdVs to indicators ratios as point estimates and as probability density functions in order to consider the variability and uncertainty of the data. Further studies will focus on verifying these ratios in larger data sets. The estimated ratios refer to HAdV genomic copies that do not correspond to infective particles. Further specific studies will be carried out, in order to clarify the relationship between biomolecular and infectivity viral data, thus filling the current gap in the scientific literature.
Funding The analytical data derived from several studies, funded by national and international institutions: Tuscany Region (Agreement for the protection of coastal-marine waters, Regional Decree No 14 of 7 October 2014) and EU (VIROBATHE-EU FP6 513648). Acknowledgement We wish to thank to English for Academics (e4ac.com) for editing and proofreading the manuscript. References Albinana-Gimenez, N., Miagostovich, M.P., Calgua, B., Huguet, J.M., Matia, L., Girones, R., 2009. Analysis of adenoviruses and polyomaviruses quantified by qPCR as indicators of water quality in source and drinking water treatment plants. Water Res. 47 (3), 2011–2019. Bambic, D., McBride, G., Miller, W., Wuertz, S., 2011. Quantification of Pathogens and Sources of Microbial Indicators for QMRA in Recreational Waters. Pathogen and Human Health. Final Report. WERF, Water Environment Research Foundation Innovation (PATH2R08 (table 6-1a)). Bofill-Mas, S., Albinana-Gimenez, N., Clemente-Casares, P., Hundesa, A., RodriguezManzano, J., Allard, A., Calvo, M., Girones, R., 2006. Quantification and stability of human adenoviruses and polyomavirus JCPyV in wastewater matrices. Appl. Environ. Microbiol. 72, 7894–7896. Calgua, B., Mengewein, A., Grunert, A., Bofill-Mas, S., Clemente-Casares, P., Hundesa, A., Wyn-Jones, A.P., López-Pila, J.M., Girones, R., 2008. Development and application of a one-step low cost procedure to concentrate viruses from seawater samples. J. Virol. Methods 153, 79–83. Carducci, A., Verani, M., 2013. Effects of bacterial, chemical, physical and meteorological variables on virus removal by a Wastewater Treatment Plant. Food Environ. Virol. 5 (1), 69–76. Carducci, A., Casini, B., Bani, A., Rovini, E., Verani, M., Mazzoni, F., Giuntini, A., 2003. Virological control of groundwater quality using biomolecular tests. Water Sci. Technol. 47 (3), 261–266. Carducci, A., Morici, P., Pizzi, F., Battistini, R., Rovini, E., Verani, M., 2008. Study of the viral removal efficiency in a urban wastewater treatment plant. Water Sci. Technol. 58 (4), 893–897. Carducci, A., Donzelli, G., Cioni, L., Federigi, I., Lombardi, R., Verani, M., 2018. Quantitative microbial risk assessment for workers exposed to bioaerosol in wastewater treatment plants aimed at the choice and setup of safety measures. Int. J. Environ. Res. Public Health 15 (7), 1490. Eregno, F.E., Tryland, I., Tjomsland, T., Myrmel, M., Robertson, L., Heistad, A., 2016. Quantitative microbial risk assessment combined with hydrodynamic modelling to estimate the public health risk associated with bathing after rainfall events. Sci. Total Environ. 548-549, 270–279. EU, 2006. Directive 2006/7/EC of the European Parliament and of the Council of 15 February 2006 Concerning the Management of Bathing Water Quality and Repealing Directive 76/160/EEC. EU, 2018. Proposal for a Directive of the European Parliament and of the Council on the Quality of Water Intended for Human Consumption (Recast). European Commission (COM(2017) 753 final 2017/0332 (COD)). Fong, T.T., Phanikumar, M.S., Xagoraraki, I., Rose, J.B., 2010. Quantitative detection of human adenoviruses in wastewater and combined sewer overflows influencing a Michigan River. Appl. Environ. Microbiol. 76 (3), 715–723. Fongaro, G., Nascimento, M.A., Rigotto, C., Ritterbusch, G., da Silva, A.D., Esteves, P.A., Barardi, C.R., 2013. Evaluation and molecular characterization of human adenovirus in drinking water supplies: viral integrity and viability assays. Virol. J. 28 (10), 166. Fuhrimann, S., Winkler, M.S., Stalder, M., Nguyen-Viet, H., Babu, M., Kabatereine, N.B., Halage, A.A., Utzinger, J., Cissé, G., Nauta, M., 2016. Disease burden due to gastrointestinal pathogens in a wastewater system in Kampala, Uganda. Microb. Risk Anal. 4, 16–28. Gerba, C.P., Castro-del Campo, N., Brooks, J.P., Pepper, I.L., 2008. Exposure and risk assessment of Salmonella in recycled residuals. Water Sci. Technol. 57, 1061–1065. Girones, R., Bofill-Mas, S., 2013. Virus indicators for food and water. In: Cook, N. (Ed.), Viruses in Food and Water: Risks, Surveillance, and Control. Woodhead Publishing Ltd, Philadelphia, pp. 483–509. Girones, R., Ferrus, M.A., Alonso, J.L., Rodriguez-Manzano, J., Calgua, B., Corrêa Ade, A., Hundesa, A., Carratala, A., Bofill-Mas, S., 2010. Molecular detection of pathogens in water—the pros and cons of molecular techniques. Water Res. 44 (15), 4325–4339. Hamza, I.A., Jurzik, L., Stang, A., Sure, K., Uberla, K., Wilhelm, M., 2009. Detection of human viruses in rivers of a densly-populated area in Germany using a virus adsorption elution method optimized for PCR analyses. Water Res. 43 (10), 2657–2668. Haramoto, E., Katayama, H., Oguma, K., Ohgaki, S., 2007. Quantitative analysis of human enteric adenoviruses in aquatic environments. J. Appl. Microbiol. 103, 2153–2159. Hernroth, B.E., Conden-Hansson, A.C., Rehnstam-Holm, A.S., Girones, R., Allard, A.K., 2002. Environmental factors influencing human viral pathogens and their potential indicator organisms in the blue mussel, Mytilus edulis: the first Scandinavian report. Appl. Environ. Microbiol. 68 (9), 4523–4533. Hewitt, J., Greening, G.E., Leonard, M., Lewis, G.D., 2013. Evaluation of human adenovirus and human polyomavirus as indicators of human sewage contamination in the aquatic environment. Water Res. 47, 6750–6761.
M. Verani et al. / Science of the Total Environment 651 (2019) 1469–1475 Howard, G., Pedley, S., Tibatemwa, S., 2006. Quantitative microbial risk assessment to estimate health risks attributable to water supply: can the technique be applied in developing countries with limited data? J. Water Health 4 (1), 49–65. Iaconelli, M., Muscillo, M., Della Libera, S., Fratini, M., Meucci, L., De Ceglia, M., Giacosa, D., La Rosa, G., 2017. One-year surveillance of human enteric viruses in raw and treated wastewaters, downstream river waters, and drinking waters. Food Environ. Virol. 9 (1), 79–88. https://doi.org/10.1007/s12560-016-9263-3. Epub 2016 Sep 28. Jurzik, L., Hamza, I.A., Puchert, W., Uberla, K., Wilhelm, M., 2010. Chemical and microbiological parameters as possible indicators for human enteric viruses in surface water. Int. J. Hyg. Environ. Health 213 (3), 210–216. Kundu, A., McBride, G., Wuertz, S., 2013. Adenovirus-associated health risks for recreational activities in a multi-use coastal watershed based on site-specific quantitative microbial risk assessment. Water Res. 47 (16), 6309–6325. La Rosa, G., Muscillo, M., 2013. Molecular detection of viruses in water and sewage. In: Cook, Nigel (Ed.), Virus in Food and Water – Risk, Surveillance and Control. Woodhead Publishing. ISBN: 9780857094308. La Rosa, G., Pourshaban, M., Iaconelli, M., Muscillo, M., 2010. Quantitative real-time PCR of enteric viruses in influent and effluent samples from wastewater treatment plants in Italy. Ann. Ist. Super. Sanita 46 (3), 266–273. La Rosa, G., Sanseverino, I., Della Libera, S., Iaconelli, M., Ferrero, V.E.V., Barra Caracciolo, A., Lettieri, T., 2017. The impact of anthropogenic pressure on the virological quality of water from the Tiber River, Italy. Lett. Appl. Microbiol. 65 (4), 298–305. Lee, S.H., Kim, S.J., 2002. Detection of infectious enteroviruses and adenoviruses in tap water in urban areas in Korea. Water Res. 36, 248–256. Lin, J., Ganesh, A., 2013. Water quality indicators: bacteria, coliphages, enteric viruses. Int. J. Environ. Health Res. 23 (6), 484–506. Linden, K.G., Thurston, J., Schaefer, R., Malley, J.P., 2007. Enhanced UV inactivation of adenoviruses under polychromatic UV lamps. Appl. Environ. Microbiol. 73, 7571–7574. Love, D.C., Rodriguez, R.A., Gibbons, C.D., Griffith, J.F., Yu, Q., Stewart, J.R., Sobsey, M.D., 2014. Human viruses and viral indicators in marine water at two recreational beaches in Southern California, USA. J. Water Health 12 (1), 136–150. Marcheggiani, S., D'Ugo, E., Puccinelli, C., Giuseppetti, R., D'Angelo, A.M., Gualerzi, C.O., Spurio, R., Medlin, L.K., Guillebault, D., Baudart-Lenfant, J., Weigel, W., Helmi, K., Mancini, L., 2015. Detection of emerging and re-emerging pathogens in surface waters close to an urban area. Int. J. Environ. Res. Public Health 12 (5), 5505–5527. McBride, G.B., Stott, R., Miller, W., Bambic, D., Wuertz, S., 2013. Discharge-based QMRA for estimation of public health risks from exposure to stormwater-borne pathogens in recreational waters in the United States. Water Res. 47 (14), 5282–5297. Myrmel, M., Lange, H., Rimstad, E., 2015. A 1-year quantitative survey of noro-, adeno-, human boca-, and hepatitis E viruses in raw and secondarily treated sewage from two plants in Norway. Food Environ. Virol. 7, 213–223. Nordgren, J., Matussek, A., Mattssond, A., Svenssone, L., Lindgren, P.E., 2009. Prevalence of norovirus and factors influencing virus concentrations during one year in a full-scale wastewater treatment plant. Water Res. 43, 1117–1125. O'Toole, J., Sinclair, M., Barker, F., Leder, K., 2014. Advice to risk assessors modeling viral health risk associated with household graywater. Risk Anal. 34 (5). https://doi.org/ 10.1111/risa.12142. Owusu-Ansah, E.D.J., Sampson, A., Amponsah, S.K., Abaidoo, R.C., Dalsgaard, A., Hald, T., 2017. Probabilistic quantitative microbial risk assessment model of norovirus from wastewater irrigated vegetables in Ghana using genome copies and fecal indicator ratio conversion for estimating exposure dose. Sci. Total Environ. 601–602, 1712–1719. Poma, H.R., Gutiérrez Cacciabue, D., Garcé, B., Gonzo, E.E., Rajal, V.B., 2012. Towards a rational strategy for monitoring of microbiological quality of ambient waters. Sci. Total Environ. 433, 98–109.
1475
Silva, H.D., García-Zapata, M.T.A., Anunciação, C.E., 2011. Why the use of adenoviruses as water quality virologic marker? Food Environ. Virol. 3, 138–140. Silverman, A.I., Akrong, M.O., Amoah, P., Drechsel, P., Nelson, K.L., 2013. Quantification of human norovirus GII, human adenovirus, and fecal indicator organisms in wastewater used for irrigation in Accra, Ghana. J. Water Health 11 (3), 473–488. Sinclair, R.G., Jones, E.L., Gerba, C.P., 2009. Viruses in recreational water-borne disease outbreaks: a review. J. Appl. Microbiol. 107 (6), 1769–1780. Skraber, S., Gassilloud, B., Gantzer, C., 2004. Comparison of coliforms and coliphages as tools for assessment of viral contamination in river water. Appl. Environ. Microbiol. 70, 3644–3649. Tanner, B.D., Brooks, J.P., Gerba, C.P., Haas, C.N., Josephson, K.L., Pepper, I.L., 2008. Estimated occupational risk from bioaerosols generated during land application of Class B biosolids. J. Environ. Qual. 37, 2311–2321. Teunis, P., Schijven, J., Rutjes, S., 2016. A generalized dose-response relationship for adenovirus infection and illness by exposure pathway. Epidemiol. Infect. 30, 1–13. Thurston-Enriquez, J.A., Haas, C.N., Jacangelo, J., Gerba, C.P., 2005. Inactivation of enteric adenovirus and feline calicivirus by chlorine dioxide. Appl. Environ. Microbiol. 71, 3100–3105. U.S. EPA, 2016. Summary of Nominations for the Fourth Contaminant Candidate List (CCL 4). Office of Water (4607M). U. S. Environmental Protection Agency, Office of Water (EPA 815-R-16-006). U.S. EPA, 2018. Five-Year Review of the 2012 Recreational Water Quality Criteria. U. S. Environmental Protection Agency, Office of Water (EPA 823-R-18-001). van Heerden, J., Ehlers, M.M., Grabow, W.O.K., 2005. Detection and risk assessment of adenoviruses in swimming pool water. J. Appl. Microbiol. 99 (5), 1256–1264. Verani, M., Di Giuseppe, G., Tammaro, C., Carducci, A., 2016. Investigating the role of Acanthamoeba polyphaga in protecting human adenovirus from water disinfection treatment. Eur. J. Protistol. 54, 11–18. Verbyla, M.E., Symonds, E.M., Kafle, R.C., Cairns, M.R., Iriarte, M., Mercado Guzmán, A., Coronado, O., Breitbart, M., Ledo, C., Mihelcic, J.R., 2016. Managing microbial risks from indirect wastewater reuse for irrigation in urbanizing watersheds. Environ. Sci. Technol. 50 (13), 6803–6813. Vergara, G.G.R.V., Rose, J.B., Gin, K.Y.H., 2016. Risk assessment of noroviruses and human adenoviruses in recreational surface waters. Water Res. 103, 276–282. Vieira, C.B., de Abreu Corrêa, A., de Jesus, M.S., Luz, S.L., Wyn-Jones, P., Kay, D., Vargha, M., Miagostovich, M.P., 2016. Viruses surveillance under different season scenarios of the Negro River Basin, Amazonia, Brazil. Food Environ. Virol. 8 (1), 57–69. WHO, 2011. Guidelines for Drinking-Water Quality. World Health Organization, Geneva, Switzerland. WHO, 2016. Quantitative Microbial Risk Assessment: Application for Water Safety Management. World Health Organization, Geneva, Switzerland. Wyer, M.D., Wyn-Jones, A.P., Kay, D., Au-Yeung, H.-K.C., Girones, R., Lopez-Pila, J., de Roda Husman, A.M., Rutjes, S., Schneider, O., 2012. Relationships between human adenoviruses and faecal indicator organisms in European recreational waters. Water Res. 46, 4130–4141. Wyn-Jones, A.P., Carducci, A., Cook, N., D'Agostino, M., Divizia, M., Fleischer, J., Gantzer, C., Gawler, A., Girones, R., Höller, C., de Roda Husman, A.M., Kay, D., Kozyra, I., López-Pila, J., Muscillo, M., Nascimento, M.S., Papageorgiou, G., Rutjes, S., Sellwood, J., Szewzyk, R., Wyer, M., 2011. Surveillance of adenoviruses and noroviruses in European recreational waters. Water Res. 45 (3), 1025–1038. Zhou, L., Echigo, S., Ohkouchi, Y., Itoh, S., 2014. Quantitative microbial risk assessment of drinking water treated with advanced water treatment process. J. Water Supply Res Technol. 63 (2), 114–123.