For plants inoculated by the drop method, the number of culturable cells declined by 4 to 5 log CFU/leaf by day 2 to an average of 10 CFU/leaf . After inoculation by spray, the average number of culturable E. coli O157:H7 ATCC 700728 cells remained higher and ranged from 100 to 1000 CFU/plant, even on day 7 of the study .Although there were significantly higher numbers of culturable EC4045 cells than strain ATCC 700728 cells at 24 h after inoculation, the culturable amounts of both strains declined to equivalently low levels within 48 h on lettuce . Real-time PCR was used to quantify total E. coli O157:H7 cells on the lettuce. The number of E. coli O157:H7 cells remained equivalent to the inoculum cell density for 7 days, regardless of the strain or the inoculation method used . The addition of PMA prior to real-time PCR amplification revealed that viable cell quantities were equal to the numbers of total and culturable cells at the time of inoculation but then declined by an average of 100- fold within the first day. For both ATCC 700278 and EC4045 strains, the number of viable cells detected using PMA real-time PCR remained above the detection limit and was significantly higher than estimated by the standard plate count method .Lettuce plants in the field were spray inoculated with E. coli O157:H7 ATCC 700278 at a density of 106 CFU/plant. E. coli O157:H7 was not detected on uninoculated lettuce plants either by culture or by real-time PCR with or without PMA. Within 2 h after application onto the lettuce, culturable cell amounts declined by 2 and 4 log for plants inoculated in the spring and summer trials, respectively . Two days after inoculation, the amounts of culturable E. coli ATCC 700278 were reduced to the lower limit of detection by plating for most plants . At that time, E. coli O157:H7 was not retrieved by plating or enrichment for 3% and 13% of the spring and summer plants, respectively. In contrast to colony enumerations of E. coli O157:H7, total cell numbers determined by real-time PCR quantification were equivalent to the inoculum levels 2 h after application to lettuce plants.

Two days after application the numbers of E. coli O157:H7 estimated by real-time PCR decreased,vertical farming supplies on average, by 2 log CFU/plant and were below the limit of detection for six out of 24 plants tested for both field trials.Accurate measurements of pathogen amounts and viability are important for risk assessments and control mechanisms aimed at preventing outbreaks of food borne illness. This study was the first to apply a culture-independent method to quantify the number of E. coli O157:H7 cells on plants inoculated with the pathogen in the field. We also compared the utility of the method for quantifying virulent and avirulent pathogen survival on laboratory-grown lettuce. The first step in the development of culture-independent methods for bacterial enumeration is to ensure that the technique is optimized for the specific organism and environmental conditions of interest. While numerous assessment methods are available, we selected PMA combined with real-time PCR because this technique offered the possibility of rapid enumeration of viable cells in a variety of environments. For E. coli O157:H7 ATCC 700728, we first identified appropriate real-time PCR primers, PMA concentrations, and PMA incubation times with a goal of selective and sensitive detection of viable E. coli O157:H7 cells in the presence of high levels of dead E. coli O157:H7 cells. Among the results was the confirmation that PMA-mediated PCR-inhibition is incomplete when the product size is less than 200 bp. Although our tests on PMA real-time PCR revealed some potential short-comings with the method, most notably the interference of high quantities of inactivated cells and the relatively high numbers of cells required for detection, these parameters were within the range found in other PCR-based studies. E. coli O157:H7 strains ATCC 700728 and EC4045 were inoculated onto leaves of potted lettuce plants and incubated in a growth chamber under temperature and low humidity conditions that favored the decline of cell numbers in trends similar to those recorded in field trials performed in the Salinas Valley. This approach differs from most other laboratory studies that examined plants exposed to high moisture and temperatures supportive of pathogen growth. Within 2 days after inoculation on lettuce plants, the number of E. coli O157:H7 ATCC 700728 cells able to form a colony on TSA declined 1000-fold when the pathogen was initially applied at a level of 6 log CFU per plant or per leaf using an aerosolized spray and declined .The numbers of culturable cells remained similar for the subsequent 5 days, suggesting that the initial events after contact with the plant are crucial for determining the survival of the organism on lettuce.

In contrast with plate counts, real-time PCR targeting of the E. coli O157:H7 ATCC 700278 and EC4045 inoculants showed that the genomic DNA of the organisms remained on the lettuce in quantities nearly equivalent to the inoculum concentrations over the duration of the 7-day experiment. This finding confirmed that both strains adhered to the leaf surfaces and the recovery method was sufficient to remove the majority of the cell inoculants from the plant. Moreover, the results indicated the presence of high numbers of intact E coli O157:H7 cells on lettuce even when very few were able to form colonies on TSA. Exposure of E. coli O157:H7 suspensions to PMA prior to realtime PCR quantification facilitated the detection of cells with an intact cell membrane, and these cells were likely viable. Application of this method to E. coli ATCC 700728 and EC4045 recovered from growth chamber lettuce indicated that a significant fraction of the pathogen inoculants were viable, even though colony-based assessments indicated otherwise. This finding was particularly evident for E. coli O157:H7 inoculated onto leaves by direct drop inoculation. While an average of 5 log CFU per leaf were viable as measured by PMA real-time PCR, less than 1 log CFU per leaf were enumerated by culturing. These results are supported by previous studies reporting that E. coli O157:H7 enters into a viable but not culturable state on lettuce leaves incubated at cold temperatures. Similar outcomes were reported for other E. coli including E. coli O104:H4 in water or after exposure to toxic concentrations of copper. Once the stress condition was removed, a fraction of the cells recovered the capacity to grow in laboratory media, thereby indicating a potential of the E. coli O1O4:H4 cells to cause human disease. The formation of VBNC bacterial cells on plants also was previously described for Listeria monocytogenes on parsley. The number of viable L. monocytogenes cells was 1 to 2 log higher than the culturable cells recovered from parsley grown in greenhouses at 20uC under low relative humidity ; growth of the VBNC cells was not restored on the plants when the RH was increased to 100%. Although we did not examine for E. coli O157:H7 recovery from the VBNC-like state,vertical lettuce tower future efforts might investigate whether those cells can recover and resume growth either under growth-conducive conditions on lettuce or after removal of the cells from the plants and prior to or after plating for viable cell enumeration.

Overall, the toxigenic strain EC4045 survived in similar quantities as ATCC 700728 on lettuce. A recent study showed that certain lineages of E. coli are more commonly associated with plants and presumably have evolved the capacity to tolerate plant associated environments better than E. coli isolated from other sites. Because we only compared two strains, subsequent investigations should examine multiple attenuated and virulent O157:H7 strains isolated from different sites for their capacity to colonize and persist on lettuce under field relevant conditions. Survival of E. coli O157:H7 on lettuce also was measured in two field studies. Culturable amounts of strain ATCC 700728 declined shortly after inoculation onto plants in the field, as we reported previously. Rates of cell decline were similar to E coli O157:H7 on lettuce in the growth chamber directly inoculated in drops with a pipette. Real-time PCR estimates of E. coli O157:H7 ATCC 700728 in lettuce leaf washes showed that this strain was present on the plants immediately after inoculation and 2 h later in quantities equivalent to the inoculum levels. Importantly, the rapid decline in culturable E. coli during the first hours after application onto plants in the field was not due to an inability to remove the organism from the lettuce or from dispersal and lack of strain attachment. Rather, it appears that the majority of the E. coli cells in the inoculum either died shortly after application or entered a VBNC state. In contrast to the growth chamber experiment results, the numbers of E. coli O157:H7 cells were below detection by real time PCR within 2 days after inoculation onto field lettuce. These findings suggest that the E. coli O157:H7 cells and genomic DNA were degraded rapidly. Environmental parameters such as solar radiation, heat, and water stress could be responsible for the differences in the stability of E. coli O157:H7 DNA in the field compared with the laboratory. Alternatively, cell maintenance might depend on other microorganisms on the leaves. Also, it is notable that different lettuce cultivars were used in the field and growth chamber studies, which may have impacted survival. The potential for cultivar-dependent effects was shown for E. coli O157:H7 on lettuce cultivars grown under axenic conditions in the laboratory.

Because the E. coli O157:H7 ATCC 700728 DNA was degraded on field-grown plants within 2 days after inoculation, it is unlikely that the organism developed a VBNC state, particularly over longer time scales . However, this possibility could not be directly addressed using the PMA realtime PCR assay in the field trials. Viable cell amounts measured by culturing and PMA real-time PCR were in agreement immediately after application of ATCC 700728 onto laboratory grown lettuce, but PMA-mediated detection was impaired on plants from the field. For those plants, the viable cell number estimates for strain ATCC 700278 were 10-fold lower as measured by PMA real-time PCR than by colony enumeration and total cell numbers estimated by real-time PCR. These differences might have been due to the higher turbidity or opacity of the lettuce plant washes from the field samples, thereby preventing light from penetrating the suspension during the PMA photo inactivation step. This interference would prevent the inactivation of free PMA, resulting in sufficient quantities of the compound to bind genomic DNA released from viable cells during the subsequent DNA extraction and amplification steps. In addition, the PMA real-time PCR assay was unable to detect low numbers of cells. Attempts to detect the ATCC 700278 strain after concentrating the leaf washes were unsuccessful . Similarly, PMA real-time PCR was found to be more reliable for viable cell detection in diluted wastewater than in pooled and concentrated wastewater samples. Such factors strongly limit the overall usefulness of this approach for field-grown plants. However, this method is informative for examination of E. coli O157:H7 on ‘‘cleaner’’ plants grown in the growth chamber and not exposed to the biotic conditions that are common outdoors. In conclusion, this study illustrates the similarities and differences between controlled studies of human pathogens on plants in a growth chamber and experiments examining the population dynamics of pathogens on plants under production-like conditions in the field. By applying relevant environmental conditions and droplet inoculation in the growth chamber, we were able to more closely mimic the rapid decline in E. coli O157:H7 culturability that was observed after inoculation of this organism onto lettuce plants in the Salinas Valley. Culture-independent assessments confirmed that the pathogen remains on the plant long after application. However, field studies showed that at least for the majority of E. coli O157:H7 cell inoculants, the loss in culturability was most likely due to cell death rather than an inability to form colonies on standard laboratory media. Hence, this work confirmed our observations that low numbers of E. coli O157:H7 persist on lettuce grown in the Salinas Valley and variations in pathogen survival among individual plants are dependent on other unknown factors .Ethnically diverse populations are disproportionately exposed to hazardous environmental materials by virtue of living in close proximity to toxic waste materials.