Deep groundwater, even if generally protected, could be contaminated by surface or rain water infiltration through soil fractures, septic tanks, cesspits, land irrigation, disposal of wastewater and disposal of muds from depuration systems. The sanitary importance of such possible contamination is related to the different uses of the water and it is at the maximum level when it is intended for human use. Routine microbiological analyses do not consider viruses, only bacterial parameters, as contamination indicators. However, it is known that enteric viruses can survive a long time in deep aquifers and that they may not always be associated with bacterial indicators. The virological analysis of waters intended for drinking use is provided only as an occasional control exercised at the discretion of the sanitary authority. Technological difficulties with obtaining data about groundwater viral contamination led to a study to devise rapid and efficient methods for their detection and the application of these methods to samples from different sources. Four acid nucleic extraction techniques have been tested (classic proteinase K- phenol/chloroform, QIAamp Viral RNA Kit (Qiagen), SV Total RNA Isolation System (Promega) and NucleoSpin Virus L (Macherey-Nagel). Sensitivity and specificity of RT-PCR protocols for entero- (EV), hepatitis A (HAV) and small round structured (SRSV) viruses have been verified. Deep groundwater samples (100 L) were concentrated (2-step tangential flow ultrafiltration) and the concentrate contaminated with serial 10-fold dilutions of a known titre of poliovirus type 3. Extracted RNA was concentrated (microcon-100) and analysed by RT-PCR using specific EV primers and visualising amplification products by agarose gel electrophoresis. In addition, two different methods of RT-PCR for non-cultivable viruses have been tested: (a) RT-PCR and nested RT-PCR for HAV and (b) RT-PCR with generic primers and RT-PCR with specific primers for SRSV. Different specificity tests have been carried out in the presence of some of the commoner microorganisms. The most efficient, sensitive and specific protocols were used to test 35x100L deep groundwater samples. Sample concentrates were split with one part treated with chloroform and analysed by cell culture (BGM and Frp/3, derived from FrHK/4, cells) and the other tested by RT-PCR for HAV, EV and SRSV. Results demonstrated the high efficiency of the classic and QIAamp methods. Microcon-100 did not increase the sensitivity of the technique used. The highest sensitivity was observed for RT-PCR with specific primers for SRSV and for nested RT-PCR for HAV. One sample showed a cytopathic effect, not confirmed at the third subculture, while the RT-PCR allowed the detection of echovirus 7. Cell culture did not allow detection of the majority of the enteric viruses while PCR gave sensitive, specific and rapid detection of a range of agents in the same samples. Even if it was impossible to fix a virological quality standard, it would be necessary to find a viral indicator in order to achieve a complete preventive check which would be particularly useful in some cases (e.g. water never used before, after pollution accidents, for seasonal checking).
Temporal variations of groundwater tables and implications for submarine groundwater discharge: a 3-decade case study in central Japan
