Occurrence of pathogenic microorganisms in the Saint Lawrence River (Canada) and comparison of health risks for populations using it as their source of drinking water.

A 300-km portion of the Saint Lawrence hydrological basin in the province of Québec (Canada) and 45 water treatment plants were studied. River water used by drinking water treatment plants was analyzed (6-L sample volumes) to determine the level of occurrence of bacterial indicators (total coliforms, fecal coliforms, and Clostridium perfringens) and pathogens (Giardia lamblia, Cryptosporidium, human enteric viruses). Pathogens and bacterial indicators were found at all sites at a wide range of values. Logistic regression analysis revealed significant correlations between the bacterial indicators and the pathogens. Physicochemical and treatment practices data were collected from most water treatment plants and used to estimate the level of removal of pathogens achieved under cold (0 degree C-4 degrees C) and warm (20 degrees C-25 degrees C) water temperature conditions. The calculated removal values were then used to estimate the annual risk of Giardia infection using mathematical models and to compare the sites. The estimated range of probability of infection ranged from 0.75 to less than 0.0001 for the populations exposed. Given the numerous assumptions made, the model probably overestimated the annual risk, but it provided comparative data of the efficacy of the water treatment plants and thereby contributes to the protection of public health.

Sources of variation in isolation rate of Giardia lamblia cysts and their homogeneous distribution in river water entering a water treatment plant.

The objective of this work was to determine if differences in the number of Giardia cysts measured in river water were due to the method itself, the analyst, or real differences in the distribution of these cysts in water. To minimize the methodological differences, centrifugation only was used as the primary concentration method. Differences were observed between results from different analysts and they were identified as technical errors. Once the method had been well established, cysts were found to be distributed homogeneously in the river water tested. Small differences were observed among samples collected sequentially at the same time, as well as for samples collected on different days or at different times on the same day. The differences reported in the literature in the number of Giardia cysts detected in water samples from the same site could be an artifact of the methods more than true differences in the counts.

Fermentation and toxin studies of the molluscicidal strains of Bacillus brevis.

Strain SS86-4 was one of 40 Bacillus brevis strains shown to be molluscicidal to the schistosomiasis snail vector Biomphalaria glabrata. When grown in mB4 medium in 2-L fermentors, SS86-4 was molluscicidal only if fructose or phenylalanine was present in the medium. This is reminiscent of secondary fermentation factor effects, in this case an antioxidant effect. In vivo proteases also were capable of reducing molluscicidal activity. The molluscicidal toxin has an LC50 of 1 microgram toxin protein ml-1 (approx. 1 p.p.m.) and may be described as a small proteinaceous, heat-stable, oxygen-sensitive entity associated with the particulate portion of the cell wall fraction of B. brevis that is formed prior to sporulation. Initial information indicates that its HPLC signature shows major peaks at 148.37 and 163.96 s and consists of two bands of approximately 5.3 kDa and 8.7 kDa on PAGE gel.