DNA-based identification of foliar microbiota with potential to predict or preclude pathogen establishment on field-grown leafy greens.

This proposal aims to gather, interpret, and translate DNA-based information from the field to help the leafy greens industry better predict disease outbreaks. The information we seek is accessible through a technology called high-throughput DNA pyrosequencing which is superior over older methods in revealing the identity and relative abundance of members that make up the microbial communities (‘microbiota’) that are naturally associated with leafy greens. We are interested in these communities because we think, based on earlier research, that they contain strains of microorganisms, particularly bacteria, which can be exploited to predict or delay the establishment in the field of human and plant pathogens. We plan to collect DNA information from a large number of leafy samples collected from different locations and field conditions. This is necessary to increase our chances of being able to pick out, from the huge number of microorganisms that make up the leaf microbiota, those that show a correlation with the occurrence of pathogens, and thus have true potential as challengers or foretellers of pathogen establishment in the field. Such microbiota-based metrics may help the industry in their efforts to assess the probability of outbreaks of food-borne illnesses or crop losses due to plant diseases.

Recurrent outbreaks of food-borne illnesses linked to the consumption of lettuce and other leafy greens produced in the California Central Coast region are of great concern to both the industry and public health regulators. The ability to predict outbreaks of Escherichia coli O157:H7, Salmonella, and other such pathogens hinges in part on the effectiveness of sampling strategies for testing leafy greens from the field for a contamination event. This proposal addresses the question whether current sampling strategies can be improved to increase the probability of finding a positive should pathogens be present. We have good evidence to support the notion that the composition of the microbial community (‘microbiota’) that is associated with pre-harvest lettuce leaf surfaces is correlated with, and may be predictive of, foliar contamination with unwanted pathogens. This evidence is based on previous culture-independent, DNA-based analysis of bacteria that naturally occurred on lettuce heads from different growing regions in the Salinas Valley region. The analysis uncovered a relationship between the relative abundance of certain bacterial taxa and that of Xanthomonas campestris, a species which includes X. campestris pv vitians (Xcv), the causative agent of the lettuce disease bacterial leaf spot. While we do not know what cause-and-effect mechanisms underlie this relationship, they offer practical utility for predicting and possibly mitigating Xcv outbreaks. Efforts to establish a similar link between the lettuce leaf microbiota and the presence of human pathogens such as Escherichia coli O157:H7 have failed so far, primarily because of a bias in our leaf sampling, in that none of the leaves were positive for contamination with E. coli O157:H7, and so a link could not be made. The research proposed here is aimed at amending this bias by targeted profiling of the microbiota on leafy greens samples from fields or plots that are suspected or confirmed positive for E. coli O157:H7 or other pathogenic bacteria. Also, the proposal seeks funding for further exploration of our Xanthomonas observation to serve as a proxy so as to establish what factors impact our ability to effectively incorporate foliar bacterial profiling of field samples into sampling strategies that are aimed at assessing the likelihood of pathogen establishment in the field. One such factor is the plant itself: recent studies have shown that different lettuce accessions carry distinctly different bacterial communities on their foliage, so that some accessions or cultivars may be better than others for the purpose of microbiota-based monitoring for pathogen establishment. We will assess this experimentally by screening lettuce cultivars for leaf microbiota that resemble in composition the ones that we identified earlier as being correlated with low/high foliar loads of Xanthomonas. This could eventually open the way for screening and breeding of lettuce cultivars with microbiota-driven resistance to the establishment of unwanted pathogens, including human pathogens.

1. Determine to what extent differences in bacterial populations on leaves between cultivars of lettuce and spinach correlate with the ability of these populations to resist invasion by pathogens, including EcO157:H7 and Xcv. 

2. Address the temporal variability in bacterial community composition by longitudinal sampling of lettuce fields in the Salinas valley.

We demonstrated clear differences in the leaf microbiota of lettuce and spinach plants growing at the same time and in the same field, suggesting that the plant is a major determinant in the outcome of bacterial colonization of plant foliage. We found partial evidence for cultivar-dependent differences in the bacterial communities on lettuce leaves, but there were no major differences among cultivars in their ability to support E. coli survival, suggesting that cultivar choice is an unlikely contributor to pathogen persistence in case of a contamination event. Overall, the decline in Escherichia sequences on E. coli-inoculated plants mirrored the decrease in culturable E. coli that could be recovered from these plants. This agreement between cultureindependent and -dependent observations suggests a rapid turnover of sprayed E. coli cells and their DNA on contaminated plants. In the course of the growing season, leaves of lettuce and spinach accumulated high levels of Enterobacteriaceae and culturable coliforms. An important question for the future is whether these contribute to an increased probability of false-positive outcomes in tests for E. coli O157:H7 and other enteropathogens.