Search

Cultures of three Listeria monocytogenes serotypes and three Salmonella spp. were applied to the exterior surfaces of waxed cardboard or plastic milk containers. Contamination sites were sampled with premoistened cotton swabs during 14 d of refrigeration. Unstressed cells of Listeria survived up to 14 d on the surfaces of waxed (1 serotype) and plastic (3 serotypes) containers. Heat-stressed cells of all three serotypes of Listeria survived for 2 d on both types of containers. One serotype survived for 4 d, but only on plastic containers. Unstressed cells of all three Salmonella strains survived up to 14 d on both types of containers. Heat-stressed Salmonella strains survived up to 2 d (waxed containers) and 4 d (plastic containers).

A strain of Yersinia enterocolitica isolated from a patient in a milk-associated yersiniosis outbreak in Tennessee, Mississippi and Arkansas in the summer of 1982 was used to contaminate the exterior of refrigerated milk containers. The bacteria survived on the containers for as long as 21 d, as shown by recovery on MacConkey agar plates or in veal infusion broth. Y. enterocolitica was not detected in milk poured from the contaminated containers.

An experimental suckling mouse intraperitoneal injection test was compared with four plasmid-associated tests (adult mouse peroral exposure, adult mouse intraperitoneal injection, auto-agglutination and plasmid detection by gel electrophoresis) to measure Yersinia enterocolitica pathogenicity. Of eight Vwa plasmid-harboring strains (O:3; O:4,32; O:5,27; O:8; O:9; O:13; O:21; and O:Tacoma) and one isogenic plasmidless strain (O:8), all Vwa plasmid-harboring strains gave identical results in all tests except the two adult mouse tests. In studies of 35 clinical strains of Y. enterocolitica recently isolated during two foodborne outbreaks, a comparison of the autoagglutination, gel electrophoresis for Vwa plasmid detection and suckling mouse tests showed that 29 strains (83%) gave identical results in all three tests. The other six strains produced different reactions in the plasmid detection and autoagglutination tests, indicating that neither test alone is sufficient to evaluate the virulence of Y. enterocolitica . To compare the sensitivity of these in vitro tests with a biological assay (the suckling mouse intraperitoneal injection test), a mixture of plasmid-harboring (P+) and plasmidless (P-) isogenic Y. enterocolitica cells was examined. The suckling mouse test was more sensitive and consistent in detecting the Vwa plasmid (as evidenced by mouse lethality). A bacterial population containing 0.1 % P + cells induced a lethal infection in the suckling mouse, whereas the other two tests required at least 10% P + cells for detection of the Vwa plasmid. The 50% lethal dose (LD50) in the suckling mouse was directly proportional to the number of Vwa-harboring cells in the culture.

From December 1981 to February 1982, 87 individuals (ages two months to 74 years) in the Seattle, WA, area developed the clinical symptoms of yersiniosis. Illness was related to consumption of commercial tofu (soybean curd) contaminated with Yersinia enterocolitica . The six Y. enterocolitica strains recovered from the hospitalized patients indicated that two antigenically distinct strains, 0:8 and 0:Tacoma, were involved. At the manufacturing site of the incriminated tofu, 112 Y. enterocolitica strains were recovered, of which two were serotype 0:8. The reactions of these strains were similar to those of clinical 0:8 strains in biochemical tests and in eight virulence factor tests. The LD50 for suckling mice was identical for all strains which killed mice. Although the causative organism(s) was not recovered from other samples of packaged tofu, our findings incriminated water used in the processing of tofu as the source of infection. The source of the second Y. enterocolitica strain (0:Tacoma) in this outbreak was not identified.

An adult mouse (18-20 g) model was developed for studying the pathogenesis of Campylobacter isolates. Iron-loaded BALB/c mice given 10(8)-10(9) Campylobacter colony forming units by intraperitoneal injection developed a severe mucoid diarrhea within 4 h. Severe diarrhea, consisting of unformed stools containing blood, mucus, and fecal leukocytes, persisted for 24 h. Diarrheal symptoms in surviving mice resolved gradually; no diarrhea was observed 5 days after inoculation. Mice not pretreated with iron developed no diarrheal symptoms, and no severe diarrhea was produced in mice inoculated orally. A transient (less than 24 h) bacteremia occurred in mice inoculated either orally or intraperitoneally. Liver, spleen, and kidney were positive for Campylobacter for 48 h; intestinal contents were positive for 5-7 days. Mice given greater than or equal to 10(10) colony forming units showed symptoms of endotoxemia (ruffled fur, inactivity, shaking, tearing, and hypothermia) and died without diarrheal symptoms. Mice given nonpathogenic Escherichia coli strain HB101, heat-killed C. jejuni cells (greater than 10(10)), C. jejuni lipopolysaccharide extract, or purified lipopolysaccharide from either Vibrio cholerae 569B or Salmonella typhimurium showed no diarrheal symptoms.

Binding of cholera toxin to Giardia lamblia was demonstrated by two slightly different methods: an immunofluorescence technique using antibody to cholera toxin and anti-rabbit immunoglobulin G conjugated to fluorescein isothiocyanate, and a one-step fluorescence method in which G. lamblia was incubated with the B subunit of cholera toxin conjugated to fluorescein isothiocyanate.

The effect of iron concentrations in culture media on supernatant yields of campylobacter cytotonic toxin (CCT) was studied. Of the 118 Campylobacter spp. strains surveyed, 78.8% produced toxin in brucella broth or in casamino acids--yeast extract (CYE) broth. When the iron concentration of CYE was increased from 0.44 microgram/mL (7.9 microM) to 0.65 microgram/mL (11.6 microM) by the addition of ferric chloride, 94.9% of the strains were positive for toxin in a ganglioside GM1 based, enzyme-linked immunosorbent assay, using antibody to affinity-purified CCT. The addition of iron as ferrous sulfate was less effective. When four toxin-positive strains were grown in a deferrated medium of conalbumin-treated CYE with 0.04-0.08 microgram iron/mL (0.72-1.43 microM), two of the culture supernatants became negative (absorbance at 410 nm, less than 0.1 and less than 10 ng CCT/mL), and two produced about 90% less CCT but were still classified as positive (absorbance, greater than or equal to 0.1 and greater than or equal to 10 ng CCT/mL). It was therefore concluded that the production of CCT by Campylobacter spp. is influenced by iron concentration.

COLLEAGUES Many animal models have been proposed for infection with Campylobacter spp. [1-6]. We have developed an animal model, using adult (18-20 g) Cox-Swiss female mice, that is inexpensive, easy to use, and representative of the human illness caused by Campylobacter. All Campylobacter strains used in this work were grown overnight with shaking in casamino acids-yeast extract (CYE) broth [7] supplemented with 10 tig of ferric chloride/ml. All cultures were incubated at 38 C in an atmosphere of 85% nitrogen, 10% carbon dioxide, and 5% oxygen; they were centrifuged at 10,000 g for 30 min and resuspended in CYE broth. Ip inoculation of mice with >106 but 106 but