Nontuberculous mycobacteria (NTM) are a major cause of opportunistic infection in immunocompromised hosts. Because there is no evidence of person-to-person transmission and NTM have been found in drinking water, the environment is considered a likely source of infection. In this study the widespread occurrence of NTM was examined in drinking water, bottled water, and ice samples. A total of 139 samples were examined for NTM by a membrane filtration culture technique followed by PCR amplification and 16S rRNA sequence determination to identify the isolates. NTM were not detected in bottled water or cisterns but were detected in 54% of the ice samples and 35% of the public drinking-water samples from 21 states. The most frequently occurring isolate was M. mucogenicum (formerly referred to as an M. chelonae-like organism).

Genotypic and phenotypic assays for glutamate decarboxylase (GAD) and ␤-D-glucuronidase (GUD) were compared for their abilities to detect various strains of Escherichia coli and to discriminate among other bacterial species. Test strains included nonpathogenic E. coli, three major groups of diarrheagenic E. coli, three other non-coli Escherichia species, and various other gram-negative and-positive bacteria found in water. The genotypic assays were performed with hybridization probes generated by PCR amplification of 670-and 623-bp segments of the gadA/B (GAD) and uidA (GUD) genes, respectively. The GAD enzymes catalyze the ␣-decarboxylation of L-glutamic acid to yield ␥-aminobutyric acid and carbon dioxide, which are detected in the phenotypic assay by a pH-sensitive indicator dye. The phenotypic assay for GUD involves the transformation of 4-methylumbelliferyl-␤-D-glucuronide to the fluorogenic compound 4-methylumbelliferone. The GAD phenotypic assay detected the majority of the E. coli strains tested, whereas a number of these strains, including all representatives of the O157:H7 serotype and several nonpathogenic E. coli strains, gave negative results in the GUD assay. Both phenotypic assays detected some but not all strains from each of the four Shigella species. A strain of Citrobacter freundii was also detected by the GUD assay but not by the GAD assay. All E. coli and Shigella strains were detected with both the gadA/B and uidA probes. A few Escherichia fergusonii strains gave weak hybridization signals in response to both probes at 65؇C but not at 68؇C. None of the other bacterial species tested were detected by either probe. These results were consistent with previous reports which have indicated that the GAD phenotypic assay detects a wider range of E. coli strains than does the GUD assay and is also somewhat more specific for this species. The genotypic assays for the two enzymes were found to be equivalent in both of these respects and superior to both of the phenotypic assays in terms of the range of E. coli strains and isolates detected.

Twenty-four randomly selected clinical and environmental Vibrio vulnificus isolates were tested for virulence in iron-overloaded mice (250 mg of iron dextran per kg of body weight). The log1o 50% lethal doses of 17 isolates were lower by >3.5 loglo units in iron-overloaded mice than in control mice. These isolates were classified as virulent. The 505% lethal doses of these virulent isolates were also lower in mice that were immunosuppressed by treatment with cyclophosphamide (150 mg/kg). Four of the seven isolates initially classified as avirulent were virulent in mice that were simultaneously iron overloaded and immunosuppressed. These isolates were classified as moderately virulent. The remaining three isolates were avirulent under all conditions. The incidence of virulent strains among clinical and environmental isolates did not differ. The virulent isolates produced high titers of hemolysin, were resistant to inactivation by serum complement, produced phenolate siderophore, and utilized transferrin-bound iron. The moderately virulent isolates differed from the virulent isolates only in their increased sensitivity to inactivation by serum complement. The avirulent isolates differed from those of the other two classes in their inability to either produce significant amounts of phenolate siderophore or utilize transferrin-bound iron. A modified agar plate diffusion method for transferrin-bound iron utilization was developed to differentiate the two classes of virulent isolates from the avirulent isolates in vitro.

The lethality of Listeria isolates was determined with normal adult mice and mice that were immunocompromised by treatment with 20 mg of carrageenan per kg. The mean 50% lethal doses (LD50s) of the pathogenic isolates were significantly lower (a = 0.05) in' the immunocompromised mice than in the untreated mice, with an average reduction of 5.8 1glo units. Ini contrast, the mean LD50s of the nonpathogenic isolates were lowèr in the immunocompromised mice by an average of only 0.4 loglo unit, a differehce that was not significant (a = 0.05). Whén immunocompromised mice were used, the LD50s of pathogenic Listeria monocytogenes isolates were lower than those of nonpathogenic L. innocua and L. seeligeri isolates by '6 loglo units and lower than those of nonpathogenic L. ivanovii isolates by .-41glo units. Pathogenic L. monocytogenes isolates could be distinguished from nonpathogenic isolates by their ability to cause deaths in immunocompromised mice in 3 days at a dose of-104 CFU per mouse. An alternative procedure using iron-overloaded mice failed to effectively differentiate pathogenic Listeria isolates.

No abstract