Prevalence of Escherichia coli O157:H7 on spinach and rocket as affected by inoculum and time to harvest

Irrigation water is an important vehicle for dissemination of human pathogens to plants. As contamination in an early stage of the production chain cannot necessarily be counteracted later, cultural measures to reduce the contamination risk need to be adopted during primary production. In a two-factorial greenhouse experiment, we studied the impact of inoculum density and the interval between irrigation and harvest on the prevalence of an inoculated gfp-tagged non-pathogenic strain of Escherichia coli O157:H7. The strain was inoculated with the irrigation water at a density of log 5.6, log 6.6 and log 7.6 CFU ml−1 into the phyllosphere of fully grown crops of rocket and spinach (BBCH 49). The crops were then harvested after 3, 24, 48 and 72 h. The introduced strain decreased exponentially in numbers within 72 h, to 49.6%, 52.6% and 50.6%, respectively, in the spinach and to 58.5%, 67.4% and 73.4% in the rocket. No differences were found in the number of the total viable count of aerobic bacteria and of Enterobacteriaceae as assessed on tryptic soy agar (TSA) and violet red bile dextrose agar (VRBD), respectively. Sequencing of the 16S rRNA genes of randomly selected isolates from VRBD were identified as Enterobacter cloaceae, Enterobacter ludwigii, Pantoea sp. and Raoultella planticola as the dominant Enterobacteriaceae species in the rocket and spinach phyllosphere. We found that cessation of irrigation for three days seems not to be an adequate sanitisation treatment to exclude the possibility of viable E. coli O157:H7 cells on spinach or rocket.