Human pathogens can contaminate leafy produce in the field by various routes. We hypothesized that interactions between Escherichia coli O157:H7 and spinach are influenced by the route of introduction and the leaf microenvironment. E. coli O157:H7 labeled with green fluorescent protein was dropped onto spinach leaf surfaces, simulating bacteria-laden raindrops or sprinkler irrigation, and survived on the phylloplane for at least 14 days, with increasing titers and areas of colonization over time. The same strains placed into the rhizosphere by soil infiltration remained detectable on very few plants and in low numbers (102 to 106 CFU/g fresh tissue) that decreased over time. Stem puncture inoculations, simulating natural wounding, rarely resulted in colonization or multiplication. Bacteria forced into the leaf interior survived for at least 14 days in intercellular spaces but did not translocate or multiply. Three spinach cultivars with different leaf surface morphologies were compared for colonization by E. coli O157:H7 introduced by leaf drop or soil drench. After 2 weeks, cv. Bordeaux hosted very few bacteria. More bacteria were seen on cv. Space and were dispersed over an area of up to 0.3 mm2. The highest bacterial numbers were observed on cv. Tyee but were dispersed only up to 0.15 mm2, suggesting that cv. Tyee may provide protected niches or more nutrients or may promote stronger bacterial adherence. These findings suggest that the spinach phylloplane is a supportive niche for E. coli O157:H7, but no conclusive evidence was found for natural entry into the plant interior. The results are relevant for interventions aimed at minimizing produce contamination by human pathogens.

The possible influence of a bile salt on production of conjugated linoleic acid (CLA) by Lactobacillus reuteri ATCC 55739 was evaluated. Cells of the lactobacilli grown in MRS broth with and without linoleic acid (LA, 0.2%) were harvested and washed. The washed cells were added to buffer containing 0.2% LA and incubated 18 h at 37 degrees C. The cells, which had been grown without LA, transformed LA into CLA (mainly c9t11-C18:2) better than did those cells grown with it. When sodium glycocholate (0.3%) was added to the washed cell suspensions, about the same level of CLA was formed as in its absence regardless of whether or not the cells had been grown in broth supplemented with free LA. Thus, glycocholate that occurs in humans did not influence production of CLA by resting cells of the lactobacilli.

Binding of sodium cholate, sodium taurocholate, and sodium glycocholate by guar gum, soluble oat fiber, xanthan gum, and inulin was studied. All soluble fibers were able to bind sodium cholate, sodium taurocholate, and sodium glycocholate from a mixture of the three, but when tested individually, the fibers bound little or no sodium cholate. In general, the fibers bound higher concentrations of conjugated bile salts than they did free bile salts. Soluble oat fiber, which bound as much or more sodium cholate and sodium glycocholate as did the other 3 fibers, was tested for its effect on deconjugation of glycocholate by Lactobacillus acidophilus and Lactobacillus casei. The presence of the fiber significantly increased (P < 0.05) deconjugation by all cultures.