Efficacy of Acetic Acid or Chitosan for Reducing the Prevalence of Salmonella- and Escherichia coli O157:H7-Contaminated Leafy Green Plants in Field Systems.

Outbreaks associated with fresh-cut leafy greens continue to occur despite efforts to implement horticultural practices that minimize introduction of enteric pathogens to the crop. The experimental trials in this study were designed to examine the efficacy of an acetic acid (AA)- and chitosan-based spray treatment, applied 1 day prior to harvest, for reducing the prevalence of Escherichia coli O157:H7 (O157) and Salmonella in field-grown leafy greens contaminated at levels detectable only through enrichment culture. Responses to the treatment solution were variable and depended on the type of leafy green (leafy lettuce, spinach, or cabbage), cultivar, pathogen, and AA concentration (0.3 to 0.7%). No significant differences in E. coli O157 prevalence were found for untreated and treated cabbage heads and spinach plants (P > 0.05). In contrast, treatment significantly affected Salmonella on 'Bravo F1' green cabbage and '7-Green' spinach (P < 0.05), with odds ratios of 2.2 and 3.3 for finding the pathogen on untreated versus treated greens, respectively. Salmonella was also 7.1 times more likely to be found on an untreated lettuce plant than on a lettuce plant sprayed with a 0.7% AA treatment solution (95% confidence interval [CI], 4.1 to 12.2; P < 0.0001). In studies addressing the efficacy of chitosan (0.1 or 0.3%), this chemical failed to reduce the prevalence of either pathogen on lettuce (P > 0.05). Similarly, spraying with 0.3% AA did not affect the prevalence of Salmonella on lettuce plants (P > 0.05); however, treatment solutions with 0.4% AA reduced the likelihood of detecting Salmonella in treated versus untreated plants by 6.6 times (95% CI, 2.1 to 20.9; P = 0.0007). After the lettuce was harvested and hand washed, consumers failed to distinguish either visually or organoleptically between untreated lettuce and lettuce sprayed with an acetic acid solution (P > 0.05). These results indicate that acetic acid could be used to reduce the microbiological risk of preharvest leafy greens.

Salmonella and Enterohemorrhagic Escherichia coli Serogroups O45, O121, O145 in Wheat Flour: Effects of Long-Term Storage and Thermal Treatments.

Salmonella and enterohemorrhagic Escherichia coli (EHEC) are of serious concern in wheat flour and its related products but little is known on their survival and thermal death kinetics. This study was undertaken to determine their long-term viability and thermal inactivation kinetics in flour. Inoculation was performed using mixtures of EHEC serogroups O45, O121, O145 and Salmonella followed by storage at room temperature (23°C) or 35°C (for Salmonella). Plate counting on tryptic soy agar (TSA) and enrichment were used to assess long-term survival. For thermal studies, wheat flour samples were heated at 55, 60, 65, and 70°C and cell counts of EHEC and Salmonella were determined by plating. The δ-values were calculated using the Weibull model. At room temperature, EHEC serovars and Salmonella were quantifiable for 84 and 112 days, and were detectable for the duration of the experiment after 168 and 365 days, respectively. The δ-values were 2.0, 5.54, and 9.3 days, for EHEC O121, O45, and O145, respectively, and 9.7 days for Salmonella. However, the only significant difference among all values was the δ-value for Salmonella and serogroup O121 (p ≤ 0.05). At 35°C, Salmonella counts declined to unquantifiable levels after a week and were not detected upon enrichment after 98 days. Heat treatment of inoculated wheat flour at 55, 60, 65, and 70°C resulted in δ-value ranges of 20.0-42.9, 4.9-10.0, 2.4-3.2, and 0.2-1.6 min, respectively, for EHEC. The δ-values for Salmonella at those temperatures were 152.2, 40.8, 17.9, and 17.4 min, respectively. The δ-values obtained for Salmonella at each temperature were significantly longer than for EHEC (p ≤ 0.05). Weibull model was a good fit to describe the thermal death kinetics of Salmonella and EHEC O45, O121 and O145 in wheat flour. HIGHLIGHTS -EHEC and Salmonella can survive for extended periods of time in wheat flour.-Long-term storage inactivation curves of EHEC and Salmonella were similar.-EHEC was more sensitive to heat than Salmonella.-Weibull model was a good fit to describe thermal death kinetics of EHEC and Salmonella.-Flour storage at 35°C may be a feasible method for microbial reduction.

Survival of Salmonella enterica and Escherichia coli O157:H7 Sprayed onto the Foliage of Field-Grown Cabbage Plants.

To reduce the number of cabbage pathogen outbreaks, it is essential to understand the fate of enteric pathogens that contaminate plants in the field. To assist in that effort, two independent trials were conducted with a red cultivar (cv. Red Dynasty) and a green cultivar (cv. Bravo F1) of field-grown cabbage ( Brassica oleracea var. capitata). In the first trial, plants with small heads were sprayed with an inoculum containing both attenuated Salmonella enterica Typhimurium and Escherichia coli O157:H7 (5.0 log CFU/mL). Initial pathogen levels (ca. 3.9 log CFU per head), determined through plate count enumeration (limit of detection was 1.3 log CFU/g), dropped precipitously such that 2 days later, they could not be detected by enrichment culture in 22 to 35% of the heads. However, subsequent declines were at a slower rate; no differences were observed between red and green cabbage heads ( P > 0.05), and heads were still positive for the pathogens 22 days after being sprayed with the inoculum. As a result, the logistic model revealed that for every 2 days contaminated cabbage heads remained in the field, the probability of finding a positive sample decreased by a factor of 1.1 (95% confidence interval from 1.0 to 1.2, P = 0.0022) and 1.2 (95% confidence interval from 1.0 to 1.4, P ≤ 0.0001) for Salmonella and E. coli O157:H7, respectively. In the second trial occurring 2 weeks later, plants with medium red or green cabbage heads were sprayed with an inoculum at a dose of 3.5 log CFU/mL. A similar decay in prevalence over time occurred for green cabbage as in trial 1; however, pathogen decline in red cabbage was less in trial 2 than in trial 1. The extended persistence of pathogens in cabbage heads exhibited in both trials infers that harvest of contaminated cabbage destined for raw consumption is risky. Additional field studies are necessary to determine whether similar pathogen fates occur in other regions or climates and to clarify the effect of the maturity of red cabbage on pathogen inactivation.

Survival and internalization of Salmonella and Escherichia coli O157:H7 sprayed onto different cabbage cultivars during cultivation in growth chambers.

BACKGROUND: Cabbage may become contaminated with enteric pathogens during cultivation. Using multiple cabbage cultivars at two maturity stages (small plants or plants with small heads) in growth chamber studies, the fate (internalization or surface survival) of Salmonella and Escherichia coli O157:H7 (0157) were examined in conjunction with any potential relationships to the plant's antimicrobial content. RESULTS: Internalized Salmonella was detected in cabbage within 24 h with prevalence ranging from 62% (16 of 26) for the 'Super Red 80' cultivar to 92% (24 of 26) for the 'Red Dynasty' cultivar. Surface survival of pathogens on small cabbage plants over nine days was significantly affected by cultivar with both pathogens surviving the most on the 'Farao' cultivar and Salmonella and O157 surviving the least on the 'Super Red 80' and 'Capture' cultivars, respectively (P < 0.05). Survival of O157 was slightly higher on cabbage heads for O157 than small plants suggesting that the maturity stage may affect this pathogen's fate. An inverse relationship existed between antimicrobial levels and the pathogen's surface survival on cabbage heads (P < 0.05). CONCLUSIONS: The fate of pathogens varied with the cabbage cultivar in growth chamber studies highlighting the potential to explore cultivar in field studies to reduce the risk of microbiological contamination in this crop. © 2019 Society of Chemical Industry.

Pre-harvest internalization and surface survival of Salmonella and Escherichia coli O157:H7 sprayed onto different lettuce cultivars under field and growth chamber conditions.

Plant genotype has been advocated to have an important role in the fate of enteric pathogens residing in lettuce foliage. This study was therefore undertaken under the premise that different pathogen responses could occur in lettuce cultivars with cultivar selection being one of several hurdles in an overall strategy for controlling foodborne pathogens on field-grown produce. Up to eight lettuce cultivars ('Gabriella', 'Green Star', 'Muir', 'New Red Fire', 'Coastal Star', 'Starfighter', 'Tropicana', and 'Two Star') were examined in these experiments in which the plants were subjected to spray contamination of their foliage with pathogens. In an experiment that addressed internalization of Salmonella, cultivar was determined to be a significant variable (P < 0.05) with 'Gabriella' and 'Muir' being the least and most likely to exhibit internalization of this pathogen, respectively. Furthermore, antimicrobials (total phenols and antioxidant capacity chemicals) could be part of the plant's defenses to resist internalization as there was an inverse relationship between the prevalence of internalization at 1 h and the levels of these antimicrobials (r = -0.75 to -0.80, P = 0.0312 to 0.0165). Internalized cells appeared to be transient residents in that across all cultivars, plants sampled 1 h after being sprayed were 3.5 times more likely to be positive for Salmonella than plants analyzed 24 h after spraying (95% CI from 1.5 to 8.2, P = 0.0035). The fate of surface-resident Salmonella and Escherichia coli O157:H7 was addressed in subsequent growth chamber and field experiments. In the growth chamber study, no effect of cultivar was manifested on the fate of either pathogen when plants were sampled up to 12 days after spray contamination of their foliage. However, in the field study, five days after spraying the plants, Salmonella contamination was significantly affected by cultivar (P < 0.05) and the following order of prevalence of contamination was observed: 'Muir' < 'Gabriella' < 'Green Star' = 'New Red Fire' < 'Coastal Star'. Nine days after spray contamination of plants in the field, no effect of cultivar was exhibited due primarily to the low prevalence of contamination observed for Salmonella (8 of 300 plant samples positive by enrichment culture) and E. coli O157 (4 of 300 plant samples positive by enrichment culture). Given the narrow window of time during which cultivar differences were documented, it is unlikely that cultivar selection could serve as a viable option for reducing the microbiological risk associated with lettuce.

Disposition of Salmonella and Escherichia coli O157:H7 following Spraying of Contaminated Water on Cucumber Fruit and Flowers in the Field.

Cucumbers are frequently consumed raw and have been implicated in several recent foodborne outbreaks. Because this item may become contaminated at the farm, it is vital to explore the fate of attenuated Salmonella Typhimurium or Escherichia coli O157:H7 sprayed onto foliage, flowers, and fruit in fields and determine whether pre- or postcontamination spray interventions could minimize contamination. After spraying cucumber plants with contaminated irrigation water (3.8 log CFU/mL of Salmonella Typhimurium and E. coli O157:H7), 60 to 78% of cucumber fruit were not contaminated because the plant's canopy likely prevented many of the underlying fruit from being exposed to the water. Subsequent exposure of contaminated cucumber plants to a simulated shower event did not appear to dislodge pathogens from contaminated foliage onto the fruit, nor did it appear to consistently wash either pathogen from the fruit. Spraying flowers and attached ovaries directly with a pathogen inoculum (4.6 log CFU/mL) initially led to 100% and 65 to 90% contamination, respectively. Within 3 days, 30 to 40% of the flowers were still contaminated; however, contamination of ovaries was minimal (≤10%), suggesting it was unlikely that internalization occurred through the flower to the ovary with these pathogen strains. In another study, both pathogens were found on a withered flower but not on the fruit to which the flower was attached, suggesting that this contaminated flower could serve as a source of cross-contamination in a storage bin if harvested with the fruit. Because pre- and postcontamination acetic acid-based spray treatments failed to reduce pathogen prevalence, the probability that fruit initially contaminated at 1.3 to 2.8 log CFU of Salmonella Typhimurium or E. coli O157:H7 per cucumber would be positive by enrichment culture decreased by a factor of 1.6 and 1.9 for Salmonella Typhimurium and E. coli O157:H7, respectively, for every day the fruit was held in the field ( P ≤ 0.0001). Hence, to reduce the prevalence of Salmonella Typhimurium on cucumbers below 5%, more than 1 week would be required.

Construction and characterization of outbreak Escherichia coli O157:H7 surrogate strains for use in field studies.

Escherichia coli O157:H7 has been the causative agent of many outbreaks associated with leafy green produce consumption. Elucidating the mechanism by which contamination occurs requires monitoring interactions between the pathogen and the plant under typical production conditions. Intentional introduction of virulent strains into fields is not an acceptable practice. As an alternative, attenuated strains of natural isolates have been used as surrogates of the virulent strains; however, the attachment properties and environmental stabilities of these attenuated isolates may differ from the unattenuated outbreak strains. In this study, the Shiga toxin (stx1, stx2, and/or stx2c) genes as well as the eae gene encoding intimin of two E. coli O157:H7 outbreak isolates, F4546 (1997 alfalfa sprout) and K4492 (2006 lettuce), were deleted. Individual gene deletions were confirmed by polymerase chain reaction (PCR) and DNA sequencing. The mutant strains did not produce Shiga toxin. The growth kinetics of these mutant strains under nutrient-rich and minimal conditions were identical to those of their wild-type strains. Attachment to the surface of lettuce leaves was comparable between wild-type/mutant pairs F4546/MD46 and K4492/MD47. Adherence to soil particles was also comparable between the virulent and surrogate pairs, although the F4546/MD46 pair exhibited statistically greater attachment than the K4492/MD47 pair (p≤0.05). Wild-type and mutant pairs F4546/MD46 and K4492/MD47 inoculated into wet or dry soils had statistically similar survival rates over the 7-day storage period at 20°C. A plasmid, pGFPuv, containing green fluorescent protein was transformed into each of the mutant strains, allowing for ease of identification and detection of surrogate strains on plant material or soil. These pGFPuv-containing surrogate strains will enable the investigation of pathogen interaction with plants and soil in the farm production environment where the virulent pathogen cannot be used.

Internalization and fate of Escherichia coli O157:H7 in leafy green phyllosphere tissue using various spray conditions.

In the past decade, leafy greens have been implicated in several outbreaks of foodborne illness, and research has focused on contamination during preharvest operations. Concerns have been raised that internalization of pathogens into the edible tissue occurs where postharvest chemical interventions would be ineffective. This study was initiated to measure the degree and fate of Escherichia coli O157:H7 internalized in the phyllosphere tissue of leafy greens when spray conditions, inoculum level, and type of leafy green were varied. Two spraying treatments were applied: (i) spraying individual spinach or lettuce leaves on plants once with a high dose (7 to 8 log CFU/ml) of E. coli O157:H7 and (ii) spraying spinach, lettuce, or parsley plants repeatedly (once per minute) with a low dose (2.7 to 4.2 log CFU/ml) of E. coli O157:H7 over a 10- to 20-min period. With the high-dose spray protocol, no significant differences in the prevalence of internalization occurred between Shiga toxin-negative E. coli O157:H7 isolates and virulent isolates (P > 0.05), implying that the Shiga toxin virulence factors did not influence internalization or the subsequent fate of those populations under these test conditions. Significantly greater internalization of E. coli O157:H7 occurred in spinach leaves compared with lettuce leaves when leaves were sprayed once with the high-dose inoculum (P < 0.05), whereas internalization was not observed in lettuce leaves but continued to be observed in spinach and parsley leaves following repeated spraying of the low-dose inoculum. Based on these results, it is surmised that a moisture film was generated when spraying was repeated and this film assisted in the mobilization of pathogen cells to plant apertures, such as stomata. E. coli O157:H7 cells that were internalized into spinach tissue using a low-dose repeat-spray protocol were temporary residents because they were not detected 2 days later, suggesting that plant-microbe interactions may be responsible.

Biotic and abiotic variables affecting internalization and fate of Escherichia coli O157:H7 isolates in leafy green roots.

Preharvest internalization of Escherichia coli O157:H7 into the roots of leafy greens is a food safety risk because the pathogen may be systemically transported to edible portions of the plant. In this study, both abiotic (degree of soil moisture) and biotic (E. coli O157:H7 exposure, presence of Shiga toxin genes, and type of leafy green) factors were examined to determine their potential effects on pathogen internalization into roots of leafy greens. Using field soil that should have an active indigenous microbial community, internalized populations in lettuce roots were 0.8 to 1.6 log CFU/g after exposure to soil containing E. coli O157:H7 at 5.6 to 6.1 log CFU/g. Internalization of E. coli O157:H7 into leafy green plant roots was higher when E. coli O157:H7 populations in soil were increased to 7 or 8 log CFU/g or when the soil was saturated with water. No differences were noted in the extent to which internalization of E. coli O157:H7 occurred in spinach, lettuce, or parsley roots; however, in saturated soil, maximum levels in parsley occurred later than did those in spinach or lettuce. Translocation of E. coli O157:H7 from roots to leaves was rare; therefore, decreases observed in root populations over time were likely the result of inactivation within the plant tissue. Shiga toxin-negative (nontoxigenic) E. coli O157:H7 isolates were more stable than were virulent isolates in soil, but the degree of internalization of E. coli O157:H7 into roots did not differ between isolate type. Therefore, these nontoxigenic isolates could be used as surrogates for virulent isolates in field trials involving internalization.

Absence of internalization of Escherichia coli O157:H7 into germinating tissue of field-grown leafy greens.

Both growth chamber and field studies were conducted to investigate the potential for Escherichia coli O157:H7 to be internalized into leafy green tissue when seeds were germinated in contaminated soil. Internalized E. coli O157:H7 was detected by enrichment in both spinach (Spinacia oleracea L.) and lettuce (Lactuca sativa L.) seedlings when seeds were germinated within the growth chamber in autoclaved and nonautoclaved soil, respectively, contaminated with E. coli O157:H7 at 2.0 and 3.8 log CFU/g, respectively. Internalized E. coli O157:H7 populations could be detected by enumeration within leafy green tissues either by increasing the pathogen levels in the soil or by autoclaving the soil. Attempts to maximize the exposure of seed to E. coli O157:H7 by increasing the mobility of the microbe either through soil with a higher moisture content or through directly soaking the seeds in an E. coli O157:H7 inoculum did not increase the degree of internalization. Based on responses obtained in growth chamber studies, internalization of E. coli O157:H7 surrogates (natural isolates of Shiga toxin-negative E. coli O157:H7 or recombinant [stx- and eae-negative] outbreak strains of E. coli O157:H7) occurred to a slightly lesser degree than did internalization of the virulent outbreak strains of E. coli O157:H7. The apparent lack of internalized E. coli O157:H7 when spinach and lettuce were germinated from seed in contaminated soil (ca. 3 to 5 log CFU/g) in the field and the limited occurrence of surface contamination on the seedlings suggest that competition from indigenous soil bacteria and environmental stresses were greater in the field than in the growth chamber. On the rare occasion that soil contamination with E. coli O157:H7 exceeded 5 log CFU/g in a commercial field, this pathogen probably would not be internalized into germinating leafy greens and/or would not still be present at the time of harvest.

Internalization of Escherichia coli O157:H7 following spraying of cut shoots when leafy greens are regrown for a second crop.

Both spinach and lettuce were grown to harvest, cut, and then regrown after spraying the cut shoots with irrigation water contaminated with Escherichia coli O157:H7. Plant tissue was collected on the day of spraying and again 2 and 14 days later for analysis of total and internalized E. coli O157:H7 populations. Internalization of E. coli O157:H7 occurred on the day of spraying, and larger populations were internalized as the level in the spray increased. Tissue repair was slow and insufficient to prevent infiltration of E. coli O157:H7; internalized E. coli O157:H7 in shoots cut 5 days prior to exposure to E. coli O157:H7-contaminated water were not significantly different from levels in shoots cut on the same day of spraying with contaminated water (P > 0.05). Two days after spraying plants with a high level of E. coli O157:H7 (7.3 log CFU/ml), levels of internalized E. coli O157:H7 decreased by ca. 2.6 and 1.3 log CFU/g in Tyee and Bordeaux spinach, respectively, whereas populations of internalized E. coli O157:H7 decreased very little (ca. 0.4 log CFU/g) in lettuce plants that had been sprayed either on the same day as cutting or 1 day after cutting. When cut plants were sprayed with irrigation water at a lower contamination level (4.5 log CFU/ml), internalized E. coli O157:H7 was not detected in either spinach or lettuce plants 2 days later and therefore would not likely be of concern when the crop was harvested.

Fate of Escherichia coli O157:H7 and Salmonella in soil and lettuce roots as affected by potential home gardening practices.

BACKGROUND: The survival and distribution of enteric pathogens in soil and lettuce systems were investigated in response to several practices (soil amendment supplementation and reduced watering) that could be applied by home gardeners. RESULTS: Leaf lettuce was grown in manure compost/top soil (0:5, 1:5 or 2:5 w/w) mixtures. Escherichia coli O157:H7 or Salmonella was applied at a low or high dose (10(3) or 10(6) colony-forming units (CFU) mL(-1) ) to the soil of seedlings and mid-age plants. Supplementation of top soil with compost did not affect pathogen survival in the soil or on root surfaces, suggesting that nutrients were not a limiting factor. Salmonella populations on root surfaces were 0.7-0.8 log CFU g(-1) lower for mid-age plants compared with seedlings. E. coli O157:H7 populations on root surfaces were 0.8 log CFU g(-1) lower for mid-age plants receiving 40 mL of water compared with plants receiving 75 mL of water on alternate days. Preharvest internalization of E. coli O157:H7 and Salmonella into lettuce roots was not observed at any time. CONCLUSION: Based on the environmental conditions and high pathogen populations in soil used in this study, internalization of Salmonella or E. coli O157:H7 into lettuce roots did not occur under practices that could be encountered by inexperienced home gardeners.

Preharvest internalization of Escherichia coli O157:H7 into lettuce leaves, as affected by insect and physical damage.

Environmental pests may serve as reservoirs and vectors of zoonotic pathogens to leafy greens; however, it is unknown whether insect pests feeding on plant tissues could redistribute these pathogens present on the surface of leaves to internal sites. This study sought to differentiate the degree of tissue internalization of Escherichia coli O157:H7 when applied at different populations on the surface of lettuce and spinach leaves, and to ascertain whether lettuce-infesting insects or physical injury could influence the fate of either surface or internalized populations of this enteric pathogen. No internalization of E. coli O157:H7 occurred when lettuce leaves were inoculated with 4.4 log CFU per leaf, but it did occur when inoculated with 6.4 log CFU per leaf. Internalization was statistically greater when spinach leaves were inoculated on the abaxial (underside) than when inoculated on the adaxial (topside) side, and when the enteric pathogen was spread after surface inoculation. Brief exposure (∼18 h) of lettuce leaves to insects (5 cabbage loopers, 10 thrips, or 10 aphids) prior to inoculation with E. coli O157:H7 resulted in significantly reduced internalized populations of the pathogen within these leaves after approximately 2 weeks, as compared with leaves not exposed to insects. Surface-contaminated leaves physically injured through file abrasions also had significantly reduced populations of both total and internalized E. coli O157:H7 as compared with nonabraded leaves 2 weeks after pathogen exposure.

Surface and internalized Escherichia coli O157:H7 on field-grown spinach and lettuce treated with spray-contaminated irrigation water.

Numerous field studies have revealed that irrigation water can contaminate the surface of plants; however, the occurrence of pathogen internalization is unclear. This study was conducted to determine the sites of Escherichia coli O157:H7 contamination and its survival when the bacteria were applied through spray irrigation water to either field-grown spinach or lettuce. To differentiate internalized and surface populations, leaves were treated with a surface disinfectant wash before the tissue was ground for analysis of E. coli O157:H7 by direct plate count or enrichment culture. Irrigation water containing E. coli O157:H7 at 10(2), 10(4), or 10(6) CFU/ml was applied to spinach 48 and 69 days after transplantation of seedlings into fields. E. coli O157:H7 was initially detected after application on the surface of plants dosed at 10(4) CFU/ml (4 of 20 samples) and both on the surface (17 of 20 samples) and internally (5 of 20 samples) of plants dosed at 10(6) CFU/ml. Seven days postspraying, all spinach leaves tested negative for surface or internal contamination. In a subsequent study, irrigation water containing E. coli O157:H7 at 10(8) CFU/ml was sprayed onto either the abaxial (lower) or adaxial (upper) side of leaves of field-grown lettuce under sunny or shaded conditions. E. coli O157:H7 was detectable on the leaf surface 27 days postspraying, but survival was higher on leaves sprayed on the abaxial side than on leaves sprayed on the adaxial side. Internalization of E. coli O157:H7 into lettuce leaves also occurred with greater persistence in leaves sprayed on the abaxial side (up to 14 days) than in leaves sprayed on the adaxial side (2 days).

Infrequent internalization of Escherichia coli O157:H7 into field-grown leafy greens.

Several sources of contamination of fresh produce by Escherichia coli O157:H7 (O157) have been identified and include contaminated irrigation water and improperly composted animal waste; however, field studies evaluating the potential for internalization of O157 into leafy greens from these sources have not been conducted. Irrigation water inoculated with green fluorescent plasmid-labeled Shiga toxin-negative strains (50 ml of 10(2), 10(4), or 10(6) CFU of O157 per ml) was applied to soil at the base of spinach plants of different maturities in one field trial. In a second trial, contaminated compost (1.8 kg of 10(3) or 10(5) CFU of O157 per g) was applied to field plots (0.25 by 3.0 m) prior to transplantation of spinach, lettuce, or parsley plants. E. coli O157:H7 persisted in the soil up to harvest (day 76 posttransplantation) following application of contaminated irrigation water; however, internalized O157 was not detected in any spinach leaves or in roots exposed to O157 during the early or late growing season. Internalized O157 was detected in root samples collected 7 days after plants were contaminated in mid-season, with 5 of 30 samples testing positive for O157 by enrichment; however, O157 was not detected by enrichment in surface-disinfected roots on days 14 or 22. Roots and leaves from transplanted spinach, lettuce, and parsley did not internalize O157 for up to 50 days in the second trial. These results indicate that internalization of O157 via plant roots in the field is rare and when it does occur, O157 does not persist 7 days later.