VirWaTest, A Point-of-Use Method for the Detection of Viruses in Water Samples

Viruses excreted by humans and animals may contaminate water sources and pose a risk to human health when this water is used for drinking, food irrigation, washing, etc. The classical fecal bacteria indicator does not always check for the presence of viral pathogens so the detection of viral pathogens and viral indicators is relevant in order to adopt measures of risk mitigation, especially in humanitarian scenarios and in areas where water-borne viral outbreaks are frequent. At present, several commercial tests allowing the quantification of fecal indicator bacteria (FIB) are available for testing at the point of use. However, such commercial tests are not available for the detection of viruses. The detection of viruses in environmental water samples requires concentrating several liters into smaller volumes. Moreover, once concentrated, the detection of viruses relies on methods such as nucleic acid extraction and molecular detection (e.g., polymerase chain reaction [PCR]-based assays) of the viral genomes. The method described here allows the concentration of viruses from 10 L water samples, as well as the extraction of viral nucleic acids at the point of use, with simple and portable equipment. This allows the testing of water samples at the point of use for several viruses and is useful in humanitarian scenarios, as well as at any context where an equipped laboratory is not available. Alternatively, the method allows concentrating viruses present in water samples and the shipping of the concentrate to a laboratory at room temperature for further analysis.


Introduction
During the first phases of any humanitarian emergency, access to clean water supplies, sanitation, and hygiene are critical for the survival of those affected. Therefore, monitoring water quality is a priority to prevent waterborne outbreaks. It is well-known that contaminated water is frequently the origin of diseases, but it is often difficult to determine the sources of viral outbreaks such as Hepatitis E virus (HEV), even with the availability of conventional laboratory methods. The control of water quality is based on the quantification of FIB 1,2, 3,4 . However, it has been extensively documented that there is no correlation between the absence of FIB and the presence of viral waterborne pathogens such as rotavirus (RoV), norovirus (NoV), or HEV 5,6 . Thus, using the water quality criteria based on FIB might result in an underestimation of risks associated with the presence of waterborne viral pathogens. The surveillance of indicator viruses, such as human adenoviruses (HAdV), or specific pathogens would be helpful in defining the exposure to viral pathogens and identifying the potential source of human infection 7,8,9,10 and in validating the efficacy of sanitation measures 11 .
Until now, the detection of viruses in these scenarios relied on skilled staff and complex logistics. VirWaTest (virwatest.org) is aimed at the development of a simple, affordable, and portable method for the concentration and subsequent detection of viruses from water samples at the point of use.
The virus concentration is based on the principle of organic flocculation of 10 L water samples, by which viruses are recovered in smaller volumes

Method development
This procedure has been developed in the Laboratory of Viruses Contaminants of Water and Food with the collaboration of GenIUL and Oxfam Intermón. It comprises of three different steps. The first one, the viral particle concentration, is an adaptation of a skimmed milk flocculation method previously described 12,17,18 . The original method was modified as to be independent of a power supply, simpler, and without centrifugation steps.
The recovery of the VirWaTest concentration method was tested in HAdV and MS2 bacteriophage-spiked groundwater samples. The viral recovery of the VirWaTest concentration and extraction method was estimated to be 3.01% to 18.02% for MS2 and 17.52% to 44.22% for HAdV. These recoveries were calculated from the values obtained by quantitative PCR during the development of the method compared to the initial concentrations of HAdV and MS2 in the water samples after spiking them with known concentrations of these viral stocks.
The VirWaTest magnetic nucleic acid extraction was compared to a commercial RNA mini kit (e.g., QIAamp Viral RNA Mini Kit), a column-based extraction method used in the laboratory, by testing 33 river and groundwater samples spiked with HAdV and MS2 and concentrated by skimmed milk flocculation. The comparison results showed that the VirWaTest method recovery was higher in 23/33 cases for HAdV, as well as for MS2 ( Table 4). A Wilcoxon test showed p-values of 0.0005569 for HAdV and 0.02791 for MS2. VirWaTest nucleic acid extraction provides significantly higher viral recoveries than the commercial one.

Detection of HAdV in environmental water samples concentrated by the VirWaTest method
To test the developed method for viral concentration in the field, the Oxfam Water, Sanitation and Hygiene (WASH) team, located in Banghi (RCA) in March 2017, collected and concentrated viruses from five well water samples.
Also, in the area of Pedernales (Ecuador), which was affected by earthquakes in 2016 and 2017, six well water samples were collected by an Oxfam WASH team in February 2017, and its viruses were concentrated by the VirWaTest concentration method. Viral concentrates from both settings were sent to the laboratory in Barcelona for VirWaTest nucleic acid extraction and viral quantification.
In Ecuador, naturally occurring HAdV were detected in six out of the six samples analyzed, with concentration values ranging from 3.27 x 10 1 to 1.80 x 10 2 GC/L, whereas one out of the five samples collected and concentrated in Banghi (RCA) tested positive for HAdV, at a concentration of 3.46 x 10 2 GC/L ( Table 5).
MS2, added to all tested samples as internal process control, was detected in all samples tested, showing that the method was correctly performed from concentration to detection.      The method has been modified for being practicable at the point of sampling without using any electric equipment except for a batteryoperated magnetic stirrer. Some other approaches based on water filtration have been adapted to the concentration of viruses and can also be performed at the point of use. However, suspended material present in water samples often clogs the filters; thus, the small volume that may be concentrated with these systems is a serious limitation.

Materials Concentration Nucleic Acid Extraction Detection
This is the first description of a method for concentrating viruses from water samples at the point of use, regardless of the turbidity of the sample. The VirWaTest concentration method allows several samples to be processed simultaneously if the appropriate material is available. Moreover, skimmed milk flocculation has been shown to be useful for bacteria and parasite concentration 19 .
The preparation of the appropriate material is crucial for performing the procedure properly. Consider the number of water samples to be analysed and prepare the reagents and material before moving to the place where the test is needed for the preparation of the reagents and material. Several samples can be processed at the same time if the appropriate amount of material is available.
Before starting the concentration of a water sample, a known concentration of a viral stock will be used to spike the sample as process control. This method uses a dried viral stock which is rehydrated with distilled water before being added to the water sample at the point of use. This is useful to rule out false-negative results at the end of the procedure and gives an indication of the performance of the method.
Viral concentrates obtained with the VirWaTest may be further tested in the field applying the VirWaTest nucleic acid extraction and detection methods or, alternatively, concentrates may be sent to a reference laboratory at room temperature. The preservative solution added to the concentrate enables viruses to remain stable for up to 2 weeks (unpublished results).
VirWaTest nucleic acid extraction is a magnetic particle-based method. It is easy and fast and allows several samples to be processed at the same time and shows equivalent and even better recovery efficiency than methods currently used for viral nucleic acid extractions. The nucleic acids may be sent to reference labs at room temperature or, if users are confident in performing molecular detection, a quantitative PCR assay can be performed at the point of use of the original water sample.
Also, if a small laboratory facility is available, the VirWaTest concentration protocol may be coupled to standard nucleic acid extraction kits that depend on a centrifuge and standard PCR-based methods that require a freezer to maintain the reagents but that use standard thermocyclers, which are less expensive than battery-operated ones.
However, since a power supply is sometimes not available, we have optimized detection assays for MS2 bacteriophages as process control and HAdV as a viral fecal indicator, and the methodology can be customized for any other virus of interest, such as hepatitis viruses, RoV, NoV, or others, to be run in the field without needing a freezer for maintenance of the reagents, nor conventional thermocyclers but only a batteryoperated one. Several battery-operated thermocyclers are commercially available. Alternatively, if a power supply is available, other conventional qPCR equipment may be used. If an eight-tube thermocycler is used, up to two nucleic acid extractions (two different samples or two replicates from the same sample) can be tested in the same PCR assay. For each nucleic acid extraction, two different quantities will be tested to rule out enzymatic inhibition originated by the sample. The adaptation is based on the previous preparation of PCR tubes by air-drying primers, probes, and standard suspensions. Several lyophilized commercial qPCR solutions exist that could be used by applying the same procedure as described here. We have described one possibility that we considered to be easy to perform.
Comparison assays performed during the development of the method showed that the VirWaTest methods of concentration, extraction, and detection are efficient for the quantification of viruses in water samples.
The limit of detection (LOD) of the described procedure is variable since the volume collected after concentration is variable. Also, the LOD can be slightly different for different viruses. If a volume of approximately 10 L is collected, the LOD for HAdV would be around 1 x 10 2 viral GC/ L. So, relatively small concentrations of HAdV can be detected by the VirWaTest method. In Pedernales (Ecuador), six out of the six samples tested presented HAdV, in concentrations close to the LOD, ranging from 3.27 x 10 1 to 1.80 x 10 2 GC/L. In Banghi (RCA), HAdV were detected in one out of the five samples analyzed, at a concentration of 3.46 x 10 2 GC/L. The presence of HAdV, as well as the presence of MS2 as the control process in the tested samples, shows the method is useful for the recovery of viral particles from water samples, even when performed by nonexperienced users.
Feedback obtained until now indicates that viral concentration is an easy procedure to perform, with no major problems when applied at the point of use of the samples, although 8 h and 5 h steps are required. It is important to note that these steps occur without any human assistance. Moreover, a faster method based on filtration is being developed as an alternative for nonturbid waters. However, flocculation seems to be, until now, the unique method that allows the concentration of samples presenting high turbidity. Viral concentrates may be then sent to any laboratory around the world at room temperature, which also makes it easier to test viruses since sampling areas do not always have good transportation services providing cooling conditions. Alternatively, the extraction and detection protocols presented here allow testing for viral detection at the point of use of the samples.
As far as we know, this is the first method reported to be useful for the concentration and testing for the presence of viruses in water samples in the field. Further efforts should be conducted to apply the procedure to the evaluation of the presence of human viral pathogens of interest in several other humanitarian crisis scenarios. Also, the user's feedback will be necessary to provide insights into the potential implementation to make the procedure friendlier.

Disclosures
The authors have nothing to disclose.