Establishing die-off rates of surrogate and sirulent EHEC-STEC strains fromsStrawberry and cilantro surfaces: time, inoculum dose and chemical intervention - proof of concept

The fresh produce industry is facing major changes in production practices due to the implementation of the FDA Food Safety Modernization Act (FSMA). Within the new rules the standards associated with water quality are among the most contested by industry associations. Current guidelines require water that will be in direct contact with the crop, to meet specific microbiological thresholds based on the 2012-EPA recreational water standards. Alternative provisions to comply with these rules have also been allowed by FDA when water cannot meet these standards. One of these options considers a microbial die-off rate of 0.5 log per day that may occur naturally between irrigation and harvest events as a safe alternative practice. Despite this potentially useful provision, there needs to be science-based information supporting this option, especially on cilantro and strawberry for which few or no further disinfection steps are commercially available after harvest and where large volumes of surface or well water are used for frost protection (strawberry) and overhead irrigation (cilantro). This research focuses on optimizing the existing knowledge in microbial die-off of avirulent and pathogenic EHEC- STEC microorganisms to determine whether the proposed microbial die-off rate is a safe farm practice to follow when using water that cannot meet the EPA microbial standards.

The die-off kinetics of virulent strains of EHEC and STEC microorganisms are not well characterized under different agricultural environments. Multiple studies have used currently available surrogates of these pathogens in an attempt to predict pathogen survival and persistence on the surface of plants, in soil and irrigation water with limited success. Major obstacles in this effort have been identified and include: 1) lack of open field environments or greenhouse facilities where researchers could make direct comparisons of the survival and persistence of these strains without compromising the health of research personnel; 2) spread and persistence of these high risk pathogens into the environment following plant/soil inoculation studies; and 3) the potential to over- or underestimate their persistence in agricultural environments. Despite these limitations, the outcomes of multiple studies have been used by FDA as part of their decision process to develop the new Produce rules (PR) within FSMA and to adopt the 2012-EPA microbial standards for recreational water as the standards for irrigation water. Current guidelines require surface water that will be used in direct contact with the edible portion of the crop to meet a rolling geometric mean of 126 CFU/100 ml and a Statistical Threshold Value of 410 CFU/100 ml of generic E. coli. Alternative provisions when surface water exceeds these standards include the use of a microbial die-off rate of 0.5 log per day that may occur naturally in the field between irrigation events and harvest or via postharvest intervention. Despite this potentially useful alternative, there is a need to evaluate this science-based data in Eastern production environments to support this option, especially when multiple crop-specific and environmental factors significantly alter targeted die-off rates of human pathogens. Strawberry and cilantro may be significantly impacted by this provision since 1) Eastern strawberry production surface waters are frequently used for frost protection close to harvest and lack further postharvest commercially available disinfection steps; and 2) frequent sprinkler irrigation events close to cilantro harvest, coupled with hand-harvesting practices and the use of hydrocooling and flume systems, impact harvestable yields, quality and safety. All these practices could increase the risk of pathogen contamination/dissemination along the supply chain. Preharvest intervention strategies for these crops could be the most reasonable and cost-effective mitigation steps that growers could adopt to reduce pathogen persistence to non-detectable levels. The use of chlorine and peroxyacetic acid via the irrigation system to treat water, and potentially provide additional protection to these crops, coupled with accurate microbial die-off rates could provide the necessary levels of control when using water that cannot meet the proposed standards. The present concept proposal will utilize BSL3 greenhouse conditions to determine and compare side-by side the die-off kinetics of surrogate and pathogenic strains of E. coli O157 and non-STEC O157. Establishing accurate die-off rates via multiple predictive models, combined with the implementation of preharvest and/or postharvest intervention strategies, including chlorine or PAA applications through the irrigation system, could provide the targeted microbial die-off rate of 0.5 log per day proposed by FSMA.

1. Determine the die-off kinetics (using multiple predictive models) of virulent and surrogate strains of E. coli O157 and non-O157 STEC inoculated to the surface of fruit and leaves of strawberry and cilantro at two inoculum concentrations (log 3 and 6 CFU/ml) with and without the application of chlorine (100 ppm) or OxiDate 2.0 (PAA) (100 ppm). 

2. Establish whether applying an interval of days between last irrigation and harvest, using a microbial die-off rate of 0.5 log per day, is a sound food safety practice for the selected crops.

Strawberry Die-off rates for EHEC, STEC and generic E. coli were strain and concentration dependent. However, all strains displayed bimodal die-off dynamics, with higher rates between 0 and 2-DPI and lower rates thereafter. All inoculated strains at log 6.0 CFU/ml displayed linear die-off rates >0.5 log CFU/day throughout the duration of the experiments. At log 3.0 CFU/ml, all inoculated strains displayed linear die-off rates <0.5 log CFU/day. However, in reality this linear die-off was not observed with any of the treatments. Similar die-off kinetics were observed for generic E. coli, pathogenic and attenuated E. coli O157:H7, and E. coli O145. Chlorine and PAA treatments had marginal to no conclusive effects on reducing the population of the inoculated pathogens 8 days after inoculation. Cilantro As with strawberry, die-off rates for EHEC, STEC and generic E. coli also were strain and concentration dependent and similarly showed bimodal die-off dynamics, with higher rates between 0 and 2-DPI and lower rates thereafter. As with strawberry, all inoculated strains at log 6.0 CFU/ml displayed linear die-off rates >0.5 log CFU/day throughout the duration of the experiments, while at log 3.0 CFU/ml, all inoculated strains displayed linear die-off rates <0.5 log CFU/day. However, in reality this linear die-off was not observed with any of the treatments. Higher numbers of positive samples after enrichment were detected on cilantro than on strawberry leaves. For cilantro, there was no discernible difference between applications of chlorine and PAA. E. coli O45 and E. coli O111 behaved differently than the other strains but were similar to each other when inoculated on both plant systems Strawberry and cilantro combined Irrespective of pathogen fitness and virulence, die-off rates after 8-DPI did not adjust to a linear model as proposed by the FSMA Produce Safety Rule, and were strain and concentration dependent. No in-field phytotoxicity was observed in either crop when chlorine or PAA were applied at 100 and 40 ppm, respectively, over a period of 4 days. Chlorine and PAA treatments had marginal to no conclusive effects on reducing the population of the inoculated pathogens 8 days after inoculation. Overall recommendation Based on our project findings the current alternative provision of applying an in-field die-off rate of 0.5 log per day within 4 consecutive days between the last irrigation event and harvest (potential cumulative die-off of 2.0 log) needs further revision and may not provide the necessary level of protection against postharvest pathogen contamination in cilantro or strawberry plants.