Introduction

Viruses cannot be completely eliminated by sewage treatment (Anastasi et al. 2008; Carducci et al. 2008; Formiga-Cruiz et al. 2005; Gabrieli et al. 1997; Laverick et al. 2004) and, dispersed into the aquatic environment, they can reach the groundwater resources (Locas et al. 2007; Lucena et al. 2006), although these sources are more protected from contamination than surface water. In many countries, waterborne-disease (WBD) outbreaks associated with groundwater pollution are reported and often the etiological agent has been identified as a virus (Blackburn et al. 2004; Liang et al. 2006; Yoder et al. 2008).

Noroviruses (NoVs), formerly known as Norwalk-like viruses, belong to the Caliciviridae family, and they are non-enveloped single-stranded RNA viruses that cause acute gastroenteritis in humans (Green et al. 2000). In industrialized countries, NoVs are the major cause of outbreaks or sporadic nonbacterial gastroenteritis, in all age-groups (Atmar and Estes 2006; Blanton et al. 2006; Buesa et al. 2002; Fankhauser et al. 2002). Lopman et al. (2003) found that the great majority of European viral outbreaks could be attributed to NoVs. Different factors contribute to the high impact of disease caused by NoVs, such as a very low infectious dose, the absence of long lasting immunity, the stability of the viruses in the environment, and the ability to be transmitted by a variety of routes (Lopman et al. 2002, 2003).

It is also considered the most common cause of foodborne and waterborne gastroenteritis outbreaks, as reported by Lynch et al. (2006) and Maunula et al. (2005). In waterborne outbreaks, a very high proportion of the population can be affected, leading from several to hundreds of cases of gastroenteritis, followed by a secondary spread and resulting in significant economic impact.

In Italy, to our knowledge, little data are available about circulating NoV strains in the water environment; to date, only four reports have described the involvement of NoVs in the etiology of outbreaks in Italy (Boccia et al. 2002; Le Guyader et al. 2006; Prato et al. 2004; Rizzo et al. 2007) and in two outbreaks the vehicle of transmission was contaminated water.

At present, NoVs as well as other viral contaminants are not routinely tested in food and drinking water because neither national nor international law requires it, and the controls are hampered by the absence of rapid and simple standardized detection methods for viral assay. In fact, monitoring viral water quality is difficult: for most human enteric viruses only a few infectious units are required to cause a disease and such low concentrations of viruses in samples usually requires their concentration from large volumes of water. The methods routinely used to detect enteric viruses in clinical specimens, such as enzyme immune assays and cell culture, are not sensitive enough. In addition, detection of viruses by cell culture is time-consuming and applicable only to cultivable viruses. The development of a reverse transcription-polymerase chain reaction test (RT-PCR) provides a rapid and sensitive detection of pathogenic enteric viruses in water at levels that could predict water safety.

In Italy, there are very few investigations on widespread of NoVs in drinking water and in groundwater in particular and the objective of this study is to assess the occurrence of NoVs in groundwater in central Italy (Latina, Latium) (Fig. 1).

Fig. 1
figure 1

Localization of the main wells and springs in Latina, a province of Latium in central Italy

Full size image

Materials and Methods

Water Samples

The geographic area investigated in this study is Latina (Latium), located in central Italy (Fig. 1). Twenty-six water samples were collected from 12 springs and 14 wells usually used as drinking water sources, and selected for their geo-hydrological and microbiological vulnerability. All samples were collected before any treatment, including chlorine disinfection.

Bacteriological Analysis

Bacteriological analysis were performed in accordance with Italian and European law (Anonymous 2001): total coliforms and Escherichia coli were enumerated according the Standard method UNI EN ISO 9308-1, membrane filtration method (Anonymous 2002), whereas Enterococci were determined by UNI EN ISO 7899-2, membrane filtration method (Anonymous 2003). The volume used for each bacteriological test was 100 ml.

Virological Analysis

Concentration of the Water Samples

A 100 l of water was filtered at a flow rate of 12–15 l min−1 through a positively charged 10-in. cartridge filter (MK type, CUNO, Meriden, Conn, USA) placed in a filter apparatus furnished with a water meter. The complete apparatus, including tubes, was cleaned and disinfected with chlorine before use, and then rinsed with sterile distilled water for 10 min. The filters were transported to the laboratory keeping the cold chain and processed within 24–48 h after collection.

Viruses were eluted by washing the cartridge for 30 min with 1,000–1,200 ml of 0.05 M glycine with 0.3% beef extract (wt/vol.) pH 9.5 (LAB-LEMCO powder-Oxoid, Italy). The pH was neutralized with 1 N HCl and the eluate was re-concentrated through an electropositive membrane 1 MDS of 4.5-cm diameter (CUNO, Meriden, Conn, USA). The absorbed viruses were eluted with 7 ml of the same elution buffer. After neutralization and chloroform treatment, the samples were ultra-centrifuged in a bench Beckman ultracentrifuge Optima TL equipped with a TLA-100 rotor, at 50,000×g for 1 h at 4°C.

The supernatant was discarded and the pellet was re-suspended in 250 μl of sterile PBS (phosphate buffered saline, Dulbecco A-Oxoid, Italy). All final concentrated samples were stored at −20°C.

Viral RNA Extraction

Total RNA was extracted from sample concentrates with TRIzol LS reagent (Life Technologies, Invitrogen). The samples were processed in accordance with the manufacturer’s instructions, and the final RNA pellet was dissolved in 15 μl sterile RNase-free water.

Molecular Analysis

A two-step RT-semi-nested PCR (RT-snPCR) was performed to detect Norovirus genogroups I and II (NoV GG I and GG II).

The RT reaction was performed in 30 μl total volume; 15 μl of total extracted RNA were denatured at 95°C for 5 min and then quick cooled on ice for 5 min, followed by the addition of 4 U of AMV, 20 U RNase Inhibitor, 0.25 mM each deoxynucleotide triphosphate, buffer for RT, and 400 ng of random hexamers (Roche, Milan, Italy). After 10 min at room temperature, the reaction mixture was incubated for 1 h at 42°C, followed by 5 min at 95°C to denature the enzymes.

The sn-PCR was performed in a final volume of 100 μl with the primers described by Hafliger et al. (1997), which allow the simultaneous detection of various NoVs, since they are located in highly conserved regions of the capsid gene for GG I and of the RNA polymerase for GG II. In the first PCR, 15 μl of cDNA were used for each NoV GG I and GG II amplification with PCR buffer, 2 U Taq DNA polymerase, 0.2 mM dNTPs, and 0.5 μM of primers; MgCl2 2.5 and 2.0 mM were used in first and second amplification, respectively. The thermal protocol of PCRs was: denaturation for 4 min at 95°C and then 40 cycles with 40 s at 95°C, 60 s at 48°C, and 60 s at 72°C. Finally, an end-extension for 7 min at 72°C was performed.

Positive and negative controls were included in all assays. NoVs were kindly obtained from C. Beuret (Spiez Laboratory, Spiez, Switzerland).

The semi-nested PCR products were analyzed for the presence of amplicons (241 bp for GG I and 203 bp for GG II) on a 2% agarose gel containing ethidium bromide (0.5 μg/ml) and visualized under UV light illumination.

All reagents and enzymes were obtained from Promega (Milan, Italy), and the primers were obtained from M-Medical (Cornaredo, Milan, Italy).

DNA Sequencing and Sequence Analysis

Positive RT-snPCR samples were confirmed by sequencing with Big Dye Terminator Cycle Sequencing Reading Reaction version 2.0-ABI Prism DNA Sequencer—Perkin Elmer. Amplicons were purified with QIAquick PCR Purification kit (QIAGEN, Milan, Italy), according to the protocol of the manufacturer.

The obtained sequences were analyzed by BLASTn at the NIH (National Institute of Health, USA) web-site (www.ncbi.nlm.nih.gov/BLAST).

Results

Overall, 4 of 26 samples (15.38%) were positive for the presence of specific amplicons (Table 1).

Table 1 Virological results by RT-semi-nested PCR
Full size table

The positive samples showed the presence of 1 NoV GG I with an homology of 95% with Nagano isolate (accession number AY 641746), 1 NoV GG II with an homology of 97% with Hokkaido isolate (accession number AB 021463), and 2 NoVs GG II with an homology of 97% with Oxford isolate (accession number AY 588005, and AY 587989). The depth of positive wells was 50, 130, and 145 m, respectively. One well was tested twice at 40 days between samples and still found to be positive for NoVs (Table 1).

All samples were negative for traditional indicators of fecal pollution: total coliforms, Escherichia coli, and Enterococci.

Discussion

It is widely believed that groundwater is generally of good quality, because microorganisms are removed or inactivated by filtration through soil passage.

In recent years, the increase of outbreaks linked to consumption of drinking water in industrialized countries has drawn greater attention to the presence of pathogenic microorganisms in water. Since 1971, a surveillance system has been in place in the U.S. for monitoring WBD outbreaks and cases of waterborne disease. In outbreaks caused by viruses, NoVs are often the cause (Blackburn et al. 2004; Liang et al. 2006; Yoder et al. 2008).

A uniform scheme for monitoring waterborne outbreaks based on common criteria does not exist within the European Union (EU), and therefore, it is impossible to evaluate the prevalence of waterborne disease outbreaks (Poullis et al. 2002). Nevertheless, occurrences of waterborne outbreaks and pathogenic microorganisms, such as bacteria, viruses, and parasites, in drinking water have been reported in different European countries (Beuret et al. 2002; Carducci et al. 2003; Divizia et al. 1993; Lucena et al. 2006; Martinelli et al. 2007; Maunula et al. 2005; Maurer and Sturchler 2000).

In Italy, about 80% of the drinking water used by the population is obtained from underground source. However, although recent outbreaks caused by water transmitted viruses have been reported, there are very few studies on the presence of NoVs in water, unlike other countries. This study has demonstrated, although the data were based on a limited number of samples, the presence of NoVs GG I and GG II, detected by RT-PCR, in groundwater destined for human use in Latina (Fig. 1). In total, 26 water samples were analyzed and specific amplicons of NoV GG I and GG II were found in four wells. It should also be emphasized that well no. 6 (Table 1), 145-m deep, was analyzed twice at 40 days between samples and still found to be positive for NoV, which is evidence of persistent pollution of water. The wells were located in karstic formations; karstic terrains provide large amounts of water, but of poor quality, as they do not create an effective filtration system (Montagna et al. 1997; Personné et al. 1998).

In our investigation, viruses were found in samples whose bacteriological parameters did not indicate a possible fecal contamination, confirming that the traditional bacteriological parameters are not valid for viruses as already reported in the literature (Gerba et al. 1979; Lucena et al. 2006).

The RT-snPCR used is not able to distinguish between infective and non-infective virus; virus infectivity was not assessed; however, the presence of viruses in water intended for human use may pose a health risk.

Further studies are needed to assess the viral contamination in groundwater by increasing the number of samples and to ascertain the source of pollution to take suitable and effective measures of environmental protection.