Understanding waterborne protozoan parasites (WPPs) diversity has important implications in public health. In this study, we evaluated a NGS-based method as a detection approach to identify simultaneously most important WPPs using 18S rRNA high-throughput sequencing. A set of primers to target the V4 18S rRNA region of WPPs such as Cryptosporidium spp., Giardia sp., Blastocystis sp., Entamoeba spp, Toxoplasma sp. and free-living amoebae (FLA) was designed. In order to optimize PCR conditions before sequencing, both a mock community with a defined composition of representative WPPs and a real water sample inoculated with specific WPPs DNA were prepared. Using the method proposed in this study, we have detected the presence of Giardia intestinalis, Acanthamoeba castellanii, Toxoplasma gondii, Entamoeba histolytica and Blastocystis sp. at species level in real irrigation water samples. Our results showed that untreated surface irrigation water in open fields can provide an important source of WPPs. Therefore, the methodology proposed in this study can establish a basis for an accurate and effective diagnostic of WPPs to provide a better understanding of the risk associated to irrigation water.

In this study, the use of skimmed milk flocculation (SMF) to simultaneously concentrate viruses, bacteria and protozoa was evaluated. We selected strains of faecal indicator bacteria and pathogens, such as Escherichia coli and Helicobacter pylori. The viruses selected were adenovirus (HAdV 35), rotavirus (RoV SA-11), the bacteriophage MS2 and bovine viral diarrhoea virus (BVDV). The protozoa tested were Acanthamoeba, Giardia and Cryptosporidium. The mean recoveries with q(RT)PCR were 66% (HAdV 35), 24% (MS2), 28% (RoV SA-11), 15% (BVDV), 60% (E. coli), 30% (H. pylori) and 21% (Acanthamoeba castellanii). When testing the infectivity, the mean recoveries were 59% (HAdV 35), 12% (MS2), 26% (RoV SA-11) and 0.7% (BVDV). The protozoa Giardia lamblia and Cryptosporidium parvum were studied by immunofluorescence with recoveries of 18% and 13%, respectively. Although q(RT)PCR consistently showed higher quantification values (as expected), q(RT)PCR and the infectivity assays showed similar recoveries for HAdV 35 and RoV SA-11. Additionally, we investigated modelling the variability and uncertainty of the recovery with this method to extrapolate the quantification obtained by q(RT)PCR and estimate the real concentration. The 95% prediction intervals of the real concentration of the microorganisms inoculated were calculated using a general non-parametric bootstrap procedure adapted in our context to estimate the technical error of the measurements. SMF shows recoveries with a low variability that permits the use of a mathematical approximation to predict the concentration of the pathogen and indicator with acceptable low intervals. The values of uncertainty may be used for a quantitative microbial risk analysis or diagnostic purposes.