Microbial quality of wild shellfish in a tropical estuary subject to treated effluent discharge

Elevated faecal indicator levels in beaches around Darwin Harbour including near effluent outfalls have led to concerns about the safety of popular, locally harvested aquatic foods. To address these concerns, the impact of treated effluent discharge on the microbial quality of aquatic foods was assessed using culture-based and molecular tools. Seawater, shellfish (mangrove snails Telescopium and Nerita balteata, the local black lip oyster Saccostrea cucullata) and mud crabs (Scylla serrata) were collected from 13 sites close (impacted) and distal (reference) to two effluent discharge locations, in creeks and coastal areas. Sampling occurred over two dry seasons and one wet season. E. coli, typical sewage related pathogens, Salmonella and norovirus, and the molecular faecal marker Bacteroides, as well as naturally occurring marine bacteria, Vibrio parahaemolyticus and Vibrio vulnificus were tested in biota. E. coli and faecal coliforms were measured in water in water. The highest concentration of faecal coliforms in water was from the Buffalo Creek discharge site. At sites located in creeks including reference creeks, the highest concentrations of faecal coliforms in water were measured in the wet season compared to the dry season, whereas in coastal sites, no or only few (<10) faecal coliforms were detected. E. coli concentrations in both snail species were significantly higher in the wet season compared to the dry season (P < 0.0001) at both the treated effluent discharge sites and reference sites with some samples exceeding the acceptable Australian and New Zealand Food Standards (ANZFS) level based on E. coli levels in bivalves. E. coli concentrations in all of the S. cucullata samples were either below the detection limit or below the ANZFS limit for E. coli. No E. coli exceedances occurred in cooked mud crabs although low E. coli levels were measured in 3 cooked mud crab samples. Only one shellfish sample (T. telescopium) was positive for Salmonella. Norovirus and Bacteroides were not detected in any of the shellfish samples tested. V. parahaemolyticus and V. vulnificus were significantly more prevalent in shellfish samples during the wet season. Twelve virulent strains of V. parahaemolyticus were detected in T. telescopium and two in N. balteata. There was no significant difference in V. parahaemolyticus prevalence between sites. V. parahaemolyticus was detected in cooked S. serrata samples from three sites in the wet season and once in the 2016 dry season. V. vulnificus was detected in both T. telescopium and N. balteata, but not in any of the S. cucullata samples. One cooked crab meat sample from a reference site was positive for V. vulnificus. The prevalence of V. vulnificus was significantly higher at the reference sites compared to the discharge or potential impact sites. The lack of correlation between E. coli and targeted faecal pathogens in biota confirm that traditional faecal markers are not suitable surrogates for faecal pollution, at least not in the tropics. The higher prevalence of E. coli, V. parahaemolyticus and V. vulnificus in biota during the wet season compared to the dry season irrespective of treated sewage influence suggests that marine snails should not be harvested and consumed from creeks during the wet season. Any future aquaculture involving shellfish should consider these data when developing appropriate shellfish quality assurance protocols and for making decisions on where to situate farms.