An official website of the United States government.
Overarching Goal: Apply landscape ecology to develop predictive models for the presence and genetic variation of foodborne pathogens persisting under agricultural and recreational land-management. <P>Objective 1: Based on an existing isolate collection, identify environmental, meteorological, host behavior and land-use factors influencing the likelihood of Listeria, Salmonella, and Escherichia coli pathogen presence in the environment to understand what parts of the landscape are hospitable or hostile to foodborne pathogens. <P>Objective 2: Quantify genetic variation in the existing isolate collection to detect the processes structuring that variation and understand how foodborne pathogens are transmitted across landscapes including potential natural and artificial barriers to transmission.<P> Objective 3: Develop and independently test predictive models resulting from objectives 1 and 2 in farm and recreational settings to identify pre-harvest intervention and control strategies. <P>While considerable information is available about environmental stress tolerance and survival of foodborne pathogens in farm environments, no explicit ecological analysis has been performed on natural pathogen populations to analyze the influences of environmental stress and land management activities on the spread and risk of produce contamination on farm landscapes. <P>By completing the proposed activities we will provide a completely new level of insight into foodborne pathogen transmission at the field level and will facilitate the development of new methods for the prevention of produce contamination and, hence, foodborne disease. This expected outcome is realistic in that such landscape analysis has already been completed by project staff examining fecal indicator bacteria.<P> This project is relevant to New York, as environmental-contamination-prone, ground-cultivated crops (e.g., cabbage, onions) are major products of New York agriculture. The ultimate goal of this research effort is to develop reliable, widely available computational tools to predict the risk of produce contamination as a map of farm lands. We expect such maps to provide guidance about the best investment of pre-harvest sampling effort to prevent foodborne disease.
Non-Technical Summary: <BR>In 2006, E. coli O157:H7 in spinach resulted in 199 reported cases of disease and 31 cases of kidney failure. Investigating scientists reported that while feral swine were the contaminating agents, the swine were infected by diverse pathways including "a common source of exposure such as water or soil". Increasing numbers of foodborne disease outbreaks are associated with produce following contamination from irrigation water, soil or interaction of animal vectors with produce. As a consequence, produce farmers are required to control and monitor foodborne pathogens (FP) on fields; whole fields sometimes cannot be harvested if samples possibly contain FP. Improved strategies to control FP in pre-harvest produce environments are thus critical to the viability of small farms. The detection, modeling and the development of strategies to prevent transmission of FP to produce is fundamentally about understanding their ecology. We propose a new approach that uses landscape ecology as the framework to address two central questions about pathogen transmission in produce: i) what farm environments are hospitable or hostile to the persistence and transmission of FP, and ii) what do the pathogens themselves (i.e., their DNA) tell us about how they move across agricultural landscapes from environmental and artificial reservoirs to fresh produce This proposal will address these fundamental unknowns about the landscape ecology of FP enabling farmers and food safety experts to ultimately predict and prevent, through good agricultural practices, food contamination and outbreaks. <P> Approach: <BR> As part of an extensive sampling effort, our research group has already collected hundreds of foodborne pathogen (FP) isolates along with meteorological, environmental and land management data from sites in New York, Florida, and Colorado. For example in NY, 735 non-agricultural samples and 439 produce field samples yielded 65 and 72 samples positive for L. monocytogenes, respectively. Landscape ecology techniques will be used to identify the rules dictating the presence and distribution of FP. A set of techniques called ordinations and classification trees can help us determine what factors are linked to the presence of FP. To understand how FP are transmitted and structured on the landscape within each genus (e.g. within Listeria), we will first use DNA-based techniques to discern genetic relationships between isolates. Then, by applying further ordinations and GIS-based spatial analysis, we can detect how the landscape, weather and environment move FP across farm and recreational landscapes. If we observe sufficient detail in the data, we will test transmission cost-models, which are used to map the most efficient physical pathway for transmission across a landscape. After integrating the results of our analysis from Objectives 1 and 2, maps will be developed for new farm and recreational landscapes using predictions based on our data. These maps will color-code areas to reflect predicted risk of produce contamination based on a classification model with remotely-sensed landscape and meteorological data as predictors. A new sampling study will be conducted to provide an initial test of where we expect to find pathogens in new study sites.