2014 Volume 20 Issue 5 Pages 961-969
Foodborne illnesses associated with the consumption of fresh produce such as raw vegetables have become a major health concern worldwide in recent years. In the present study, we investigated the possible routes of Salmonella contamination in leafy lettuce via soil during cultivation. After 10-week cultivation of lettuce plants in soils inoculated with S. Enteritidis expressing green fluorescent protein (SE-EGFP), the bacterium was detected in soil inoculated with >104 cfu/g and from most lettuce leaves cultivated in soils inoculated with >4.4 × 107 cfu/g. As Salmonella was not detected in intact lettuce leaves or lettuce leaves with root injury cultivated in highly contaminated soils and after surface disinfection, the lettuce plants were not considered to internalize the bacterium. Overhead irrigation led to the contamination of one in 10 lettuce plants; however, all sets of three leaves of the plant were contaminated (>110 MPN/g). In an effort to prevent Salmonella contamination from soils, we investigated the effects of mulch on contamination levels during cultivation. Mulch effectively reduced Salmonella contamination levels of lettuce plants cultivated in highly contaminated soils.
The increasing consumption of raw and minimally processed fruits and vegetables, such as packaged and precut salads, etc., has been accompanied by an increase in the number of outbreaks of foodborne illnesses associated with fresh produce in developed countries (Sivapalasingam et al., 2004; Doyle and Erickson, 2008). These outbreaks have raised interest in identifying pre- and/or post-harvest sources of contamination of raw and minimally processed fruits and vegetables with pathogens.
Fresh produce such as tomato, spinach, and lettuce are reported to be vectors of pathogens (Doyle and Erickson, 2008). Moreover, leafy vegetables are often contaminated with pathogenic bacteria. For instance, outbreaks of Salmonella Typhimurium and Escherichia coli O157:H7 in Europe and USA were linked to leafy vegetables such as lettuce and spinach (Horby et al., 2003; Heaton and Jones, 2008; Parker et al., 2012). Epidemic survey also showed a high possibility of leafy vegetables infected with E. coli O157:H7 at the farm (Parker et al., 2012; Cooley et al., 2007). In particular, Parker et al. (2012) reported that a large outbreak of E. coli O157:H7 in 2006 was linked to the consumption of ready-to-eat baby spinach, likely contaminated by soil and water. From those reports, it is considered that the cultivation period is a critical point in preventing contamination with pathogenic bacteria.
The use of partially decomposed compost, the mediation by insects or wild animals, and irrigation water are possible contamination sources of vegetables during cultivation (Beuchat, 1996). The use of partially decomposed compost can be easily restricted to avoid contamination. However, irrigation water, while essential for cultivation, is considered an important contamination source of leafy vegetables. In Japan, mulch is widely used for the cultivation of leafy vegetables to prevent damage caused by cold, frost, heat, or drought, and to prevent weeds. As mulch can prevent direct contact between vegetables and the soil, mulching is expected to be effective in preventing bacterial contamination in leafy vegetables. However, there are few reports regarding the effects of mulch on contamination levels of pathogenic bacteria in vegetables.
We previously showed that Salmonella survived in soil > 70 days during tomato cultivation (Mishima et al., 2012). Once the soil is contaminated with bacterial pathogens, the pathogens can remain in the soil during cultivation and contaminate the leafy vegetables. In the present paper, we investigated several possible routes of Salmonella contamination in lettuce plants via soil during cultivation. Moreover, the use of mulch in preventing Salmonella contamination was also investigated.
Bacterial strains and culture conditions Salmonella enterica serovar Enteritidis IFO3313 was obtained from the National Institute of Technology and Evaluation-Biological Resource Center (Kisarazu, Japan) and electrophoretically transformed with a plasmid vector, pEGFP (BD Biosciences, San Jose, CA, USA), carrying the enhanced green fluorescent protein (EGFP) gene. The transformed strain was designated as SE-EGFP. One loopful of SE-EGFP was inoculated in 10 mL of Tryptic Soy Broth (Difco Laboratories Inc., Cockeysville, MD, USA) supplemented with 50 μg/mL of ampicillin (Nacalai Co. Ltd., Tokyo, Japan) and incubated with shaking at 37°C overnight. Cells were harvested by centrifugation at 5,000 × g for 5 min, resuspended in sterile water, and then serially diluted with sterile water to concentrations of approximately 100, 101, 106, 107, 108, or 109 cfu/mL.
Preparation of SE-EGFP-contaminated soil Commercially available soil (peat-moss/perlite/vermiculite) was purchased from Oishi Bussan Co. Ltd. (Tokyo, Japan). The soil was sterilized by autoclaving at 121°C for 20 min before inoculation. Approximately 100 g of soil was mixed with 50 mL of SE-EGFP at concentrations of 100, 101, 106, 107, 108, or 109 cfu/mL. The SE-EGFP concentrations in contaminated soil were approximately 10−1, 100, 105, 106, 107, or 108 cfu/g after mixing. Plastic pots 7.5 cm in diameter were filled with 45 g of contaminated soil for lettuce cultivation.
Seed disinfection Seeds of Lactuca sativa var. crispa (Banshu RedFire) were obtained from Takii Seeds Co. Ltd. (Kyoto, Japan). The seeds were surface disinfected with 70% ethanol for 10 – 15 s, and then immersed in 0.2% sodium hypochlorite solution for 10 min. The seeds were then rinsed three times with sterile water.
Growth conditions of leafy lettuce Disinfected seeds were planted in pots with contaminated soil, and were grown at 20°C with a 16-h light and 8-h dark photoperiod in a growth chamber (TOMY SEIKO Co. Ltd., Tokyo, Japan). Thirty mL of sterile tap water was added to the soil around the root daily. Furthermore, 30 mL of nutrient solution (0.2% Hyponex; Hyponex Japan Co. Ltd., Osaka, Japan) was applied to the lettuce plant once per week.
Mulching Two weeks after seed germination, the cultivation pot was covered with disinfected polypropylene film with a hole (3 – 4 cm diameter) during cultivation. Disinfection was carried out by spraying the film with 70% ethanol.
Generation of highly contaminated soil by addition of SE-EGFP suspension To examine the effect of high SE-EGFP contamination level in soil, the bacterial culture was suspended in sterile water at a concentration of approximately 109 cfu/mL and the suspension was used as irrigation water. To maintain an SE-EGFP concentration of 107 – 108 cfu/g, 30 mL of contaminated water (109 cfu/mL) was added directly to the soil around the root, and without leaf contact, once every two weeks.
SE-EGFP contamination via artificial root injury To investigate whether root injury promotes SE-EGFP invasion via the plant vasculature, the main root of the lettuce plant, which had been cultivated in SE-EGFP contaminated soil (108 CFU/g), was severed. The root was severed 1 cm below the soil surface after 8 weeks of cultivation, and the lettuce plant was subsequently replanted in SE-EGFP-contaminated soil at approximately 107 CFU/g. Lettuce leaves were harvested 2 to 9 weeks after replantation. To investigate the internalization of SE-EGFP in lettuce leaves, samples were surface disinfected as described below.
Sampling methods for soils and lettuce plants During cultivation, 5 g soil samples were analyzed for SE-EGFP levels biweekly. Lettuce plants were harvested after the indicated cultivation periods. For microbial analysis, three sections of the outermost leaves from plants were used as a sample set.
Surface disinfection of lettuce leaves To determine whether SE-EGFP was internalized by lettuce or attached to the lettuce surface, harvested lettuce leaves were surface-disinfected with 100 mL of 70% ethanol for 1 min and 100 mL of 10% sodium hypochlorite for 1 min, at room temperature. Then, the leaves were rinsed three times in 100 mL of sterile water at room temperature for 1 min.
Homogenization of samples Soil or lettuce leaf samples were transferred to Stomacher bags and ten-fold diluted (one part lettuce to 9 parts buffered peptone water) with buffered peptone water (BPW) (Nissui Pharmaceutical Co. Ltd., Tokyo, Japan). The soil and lettuce samples were then homogenized for 30 s using a Pulsifier (Microgen Bioproducts Ltd., Camberley, Surrey, UK) and a Masticator (IUL instruments, Barcelona, Spain), respectively.
Determination of viable bacterial counts and Salmonella counts One milliliter of each homogenized sample was serially diluted with sterile water and spread (100 µL) onto tryptic soy agar (TSA, Japan BD Company, Tokyo, Japan) containing 50 µg/mL ampicillin (TSA/Amp), deoxycholate-hydrogen sulfide-lactose agar (DHL, Nissui Pharmaceutical) in duplicate and incubated at 37°C for 48 h. All colonies on TSA were counted as total viable bacterial counts; colonies expressing green fluorescence on TSA/Amp were counted as SE-EGFP counts; and red colonies were also counted as Enterobacteriaceae, black colonies on DHL were counted as Salmonella counts. Viable counts were expressed as colony forming units (cfu)/g.
To measure viable counts below the lower limit of detection of the plating method, the Most Probable Number (MPN) method was applied. In this method, 10, 1, and 0.1 mL of homogenates were inoculated into each 3 test tubes with 10 mL BPW and incubated at 37°C for 20 ± 2 h. The culture (100 µL) was inoculated into 10 mL Rappaport-Vassiliadis broth (RV broth, Merck Ltd., Tokyo, Japan), and incubated at 42°C for 22 ± 2 h. Finally, one loopful of the culture was withdrawn with a sterile plastic microloop, streaked onto DHL agar, and incubated at 37°C for 24 ± 2 h. MPN values for Salmonella were determined using the MPN table and expressed as MPN/g.
Survival of SE-EGFP in soil To investigate the survival of SE-EGFP in soil, SE-EGFP-inoculated soil was examined biweekly during the 70-day cultivation period. Figure 1 shows SE-EGFP survival in soil during lettuce cultivation. SE-EGFP soil levels were observed at an inoculum level-dependent manner during cultivation. The total viable counts of SE-EGFP-inoculated soil at 101, 104, and 106 cfu/g increased up to 108 cfu/g on day 14 and the counts remained more or less steady during the remainder of the cultivation period. Soil counts of SE-EGFP inoculated at 104 and 106 cfu/g increased to about 107 cfu/g on day 14 and decreased gradually to 105 cfu/g during the remainder of the cultivation period. At an initial inoculation of 101 cfu/g, SE-EGFP counts also increased to 105 cfu/g on day 28, then decreased gradually to 103 cfu/g on day 42, and were maintained until the end of the cultivation period. However, at an initial SE-EGFP inoculation of 108 cfu/g, counts decreased gradually to 106 cfu/g during the 70-day cultivation period.
Survival of SE-EGFP in soil during lettuce cultivation. Commercially available horticultural soil was inoculated with SE-EGFP at a final concentration of 101, 104, 106, 108 cfu/g. Leaf lettuces were cultivated as described in the text. Open circles and closed circles indicate total viable counts and SE-EGFP counts, respectively.
Effects of soil SE-EGFP inoculum level on contamination of lettuce leaves To investigate the relationship between initial soil SE-EGFP level and lettuce leaf contamination, whole lettuce was examined for viable bacterial counts and Salmonella counts at harvest after 10-week cultivation. As shown in Table 1, soil viable counts ranged from 107 to 108 cfu/g independent of the initial inoculum size. Salmonella was detected in the soil at about 103 to 106 cfu/g, depending on the initial Salmonella inoculation level. Viable counts of whole lettuce ranged from 103 to 105 cfu/g regardless of the initial soil Salmonella level. However, Salmonella counts in the whole lettuce samples were less than the lower limit of detection in both the plating (<102 cfu/g) and MPN (<0.3 MPN/g) methods.
| Initial SE-EGFP inoculum (cfu/g) | Sample No. | Viable counts in soil (cfu/g) | Viable counts of whole lettuce (cfu/g or MPN/g) | |||
|---|---|---|---|---|---|---|
| Total | Salmonella | Total | Salmonella | |||
| Plating method | MPN | |||||
| 0 | 1 | 3.78 × 107 | <102 | 4.23 × 104 | <102 | <0.3 |
| 0.42 | 3 | 2.52 × 107 | 1.52 × 103 | 8.76 × 105 | <102 | <0.3 |
| 0.69 | 2 | 2.18 × 107 | 3.39 × 103 | 8.49 × 103 | <102 | <0.3 |
| 1.35 | 1 | 2.87 × 107 | 2.53 × 103 | 6.10 × 105 | <102 | <0.3 |
| 1.35 | 1 | 2.81 × 107 | 4.42 × 105 | 1.54 × 103 | <102 | <0.3 |
| 2.7 × 104 | 1 | 2.81 × 107 | 4.42 × 105 | 1.54 × 103 | <102 | <0.3 |
| 2 | <102 | <0.3 | ||||
| 3.1 × 105 | 1 | 2.99 × 107 | 7.15 × 105 | 4.46 × 103 | <102 | <0.3 |
| 2 | <102 | <0.3 | ||||
| 3.8 × 106 | 1 | 4.25 × 107 | 3.14 × 106 | 7.84 × 103 | <102 | <0.3 |
| 2 | <102 | <0.3 | ||||
| 4.4 × 107 | 1 | 4.23 × 107 | 7.03 × 106 | 2.35 × 104 | <102 | <0.3 |
| 2 | - | <0.3 | ||||
| 3 | <102 | <0.3 | ||||
| 12.1 × 108 | 1 | 1.08 × 108 | 7.20 × 106 | 2.38 × 104 | - | <0.3 |
| 2 | <102 | <0.3 | ||||
To confirm the absence of Salmonella in lettuce leaves, sets of 3 outermost leaves from lettuce were subjected to determination of Salmonella counts at harvest after 10-week cultivation. As shown in Table 2, Salmonella was not detected in every set of lettuce leaves cultivated in soil spiked at <2.7 × 104 cfu/g. Salmonella was detected at 0.36 and 0.91 MPN/g from the sets of 1st – 3rd and 7th – 9th outermost leaves, respectively, in one out of 3 lettuce plants cultivated in soil initially inoculated with levels of SE-EGFP at 3.1 × 105 cfu/g. In addition, Salmonella was detected in almost all of the sample sets, from 1st to 15th outermost leaves of one of two lettuces cultivated in the soil spiked at 4.4 × 107 cfu/g. At an initial soil contamination level of 2.1 × 108 cfu/g, Salmonella was detected from the 1st – 15th outermost leaves from one lettuce.
| Initial SE-EGFP inoculum (cfu/g) | Sample No. | Salmonella counts of leaves (MPN/g) | ||||||
|---|---|---|---|---|---|---|---|---|
| Outermost leaves | ||||||||
| 1st–3rd | 4th–6th | 7th–9th | 10th–12th | 13th–15th | 16th–18th | 19th–21st | ||
| 0 | 2 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 |
| 0.42 | 6 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - | - |
| 0.69 | 5 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - | - |
| 1.35 | 4 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - | - |
| 2.7 × 104 | 3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - |
| 3.1 × 105 | 3 | 0.36 | <0.3 | 0.91 | <0.3 | <0.3 | <0.3 | - |
| 4 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - | |
| 5 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - | - | |
| 3.8 × 106 | 3 | 2 | <0.3 | 46 | 9.3 | <0.3 | <0.3 | - |
| 4.4 × 107 | 4 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | 2.1 |
| 5 | 0.93 | <0.3 | 0.3 | 0.36 | 24 | - | - | |
| 2.1 × 108 | 4 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | <0.3 | - |
| 5 | 2.3 | 0.3 | 45 | 1.5 | 24 | <0.3 | <0.3 | |
Effect of mulching on Salmonella contamination of lettuce plants cultivated in contaminated soils To investigate the effects of mulching on contamination control, soils spiked at 105, 106, 107, or 108 cfu/g were used for lettuce cultivation with mulching. After 10-week cultivation, Salmonella levels in sets of 3 outermost leaves from lettuce were investigated (Table 3). Salmonella levels from all samples were below the detection limit (0.3 MPN/g), suggesting that mulching reduced contact between the lettuce and soil, and consequently reduced the contamination of lettuce with Salmonella.
| Initial SE-EGFP count (cfu/g) | Outermost leaves | Salmonella positive plants/Total plants | Salmonella counts in leaves (MPN/g) |
|---|---|---|---|
| 103 | 1st–3rd | 0/3 | <0.3 |
| 4th–6th | 0/3 | <0.3 | |
| 7th–9th | 0/3 | <0.3 | |
| 10th–12th | 0/3 | <0.3 | |
| 13th–15th | 0/3 | <03 | |
| 106 | 1st–3rd | 0/3 | <0.3 |
| 4th–6th | 0/3 | <0.3 | |
| 7th–9th | 0/3 | <0.3 | |
| 10th–12th | 0/3 | <0.3 | |
| 13th–15th | 0/3 | <03 | |
| 107 | 1st–3rd | 0/3 | <0.3 |
| 4th–6th | 0/3 | <0.3 | |
| 7th–9th | 0/3 | <0.3 | |
| 10th–12th | 0/3 | <0.3 | |
| 13rd–15th | 0/2 | <03 | |
| 108 | 1st–3rd | 0/3 | <0.3 |
| 4th–6th | 0/3 | <0.3 | |
| 7th–9th | 0/3 | <0.3 | |
| 10th–12th | 0/3 | <0.3 | |
| 13th–15th | 0/3 | <0.3 |
Effect of mulching on Salmonella contamination of lettuce plants cultivated in highly contaminated soil The effect of mulching on Salmonella contamination in lettuce plants cultivated in highly contaminated soil was investigated. The soil was spiked with SE-EGFP at 108 cfu/g and then, to maintain SE-EGFP soil contamination levels in the range of 106 – 108 cfu/g, 30 mL of a SE-EGFP suspension (109 cells/mL) was added to the soil biweekly during cultivation. After 10 – 17 weeks of cultivation, lettuce plants were harvested and divided to several sets of 3 outermost leaves for determination of Salmonella counts. Viable Salmonella counts of the leaves were measured by the MPN method before and after surface disinfection (Table 4). In non-disinfected lettuce plants cultivated without mulch, Salmonella was detected in 7, 5, 3, and 2 out of 8 plants for the set of 1st – 3rd, 4th – 6th, 7th – 9th, and 10th – 12th outermost leaves, respectively. The contamination levels of outer leaves were higher than those of inner leaves. The detected levels were in the range of 0.73 to >110 MPN/g in all sets from the 1st – 12th outermost leaves. On the other hand, in the case of mulched non-disinfected lettuce plants, Salmonella was detected in 3 and 2 out of 7 plants for the sets of 1st – 3rd and 4th – 6th outermost leaves, respectively. These results show that mulching can effectively reduce contamination of lettuce plants. In regard to surface disinfected samples, Salmonella was not detected in 35 samples regardless of mulching, except for one plant. Even under cultivation in highly contaminated soils, the possibility of internalization of Salmonella in lettuce plants was thought to be low.
| Mulching | Outermost leaves | Before disinfection | After disinfection | ||
|---|---|---|---|---|---|
| Salmonella positive plants/Total plants | Salmonella counts of leaves (MPN/g) | Salmonella positive plants/Total plants | Salmonella counts of leaves (MPN/g) | ||
| No | 1st–3rd | 7/8 | 0.91->110 | 0/8 | <0.3 |
| 4th–6th | 5/8 | 0.73->110 | 0/8 | <0.3 | |
| 7th–9th | 3/8 | 2.3->110 | 0/8 | <0.3 | |
| 10th–12th | 2/8 | 24->110 | 0/8 | <0.3 | |
| 13th–15th | 0/2 | <0.3 | 0/2 | <0.3 | |
| Yes | 1st–3rd | 3/7 | 0.36->110 | 0/27 | 0.3 |
| 4th–6th | 2/7 | 2.3 | 0/27 | <0.3 | |
| 7th–9th | 0/7 | <0.3 | 0/27 | <0.3 | |
| 10th–12th | 0/6 | <0.3 | 0/27 | <0.3 | |
| 13th–15th | 0/1 | <0.3 | 0/3 | <0.3 | |
Effect of root injury on Salmonella internalization of lettuce in highly contaminated soil Table 5 shows the effect of root injury on Salmonella internalization in lettuce tissues cultivated in soil spiked with SE-EGFP at 106 – 108 cfu/g. Salmonella was detected in the range from 0.3 to >110 MPN/g on sets of 3 outermost leaves of 6 out of 8 plants. The Salmonella contamination level of lettuce was lower in the inner leaves compared with the outer leaves. However, Salmonella was not detected in any sets of 3 outermost leaves of 8 plants after surface disinfection. These results indicate that root injury does not affect the internalization of Salmonella and soil contamination of lettuce occurs only on the plant surfaces, not internally via the plant vasculature.
| Outermost leaves | Before disinfection | After disinfection | ||
|---|---|---|---|---|
| Salmonella positive plants/Total plants | Salmonella counts of leaves (MPN/g) | Salmonella positive plants/Total plants | Salmonella counts of leaves (MPN/g) | |
| 1st–3rd | 6/8 | 0.3->110 | 0/8 | <0.3 |
| 4th–6th | 5/8 | 0.91->110 | 0/8 | <0.3 |
| 7th–9th | 5/8 | 0.91->110 | 0/8 | <0.3 |
| 10th–12th | 4/8 | 0.36-110 | 0/8 | <0.3 |
| 13th–15th | - | - | 0/1 | <0.3 |
Effect of overhead irrigation on Salmonella contamination of lettuce cultivated in contaminated soil To study the effect of water splash on SE-EGFP contamination of lettuce cultivated in highly contaminated soil, to which contaminated water (107 – 108 cfu/mL) was added biweekly in addition to initial inoculation, overhead irrigation with sterilized water was conducted daily during cultivation. Concurrently, the effect of mulching was also investigated. After 10 – 15 weeks of cultivation, lettuce plants were harvested. As shown in Table 6, without mulching, Salmonella was detected in all sets of at least one plant at >100 MPN/g. Even the innermost leaves (the 22nd – 24th) were contaminated with Salmonella. On the other hand, in the case of mulching cultivation, Salmonella was detected from only one set (4th – 6th) of leaves at 4.3 MPN/g. These results indicate that overhead irrigation causes Salmonella contamination in cultivated lettuce and that mulching can effectively reduce contamination levels in lettuce.
| Mulching | Outermost leaves | Salmonella positive plants/Total plants | Salmonella counts of leaves (MPN/g) |
|---|---|---|---|
| No | 1st–3rd | 2/10 | 0.3->110 |
| 4th–6th | 1/10 | >110 | |
| 7th–9th | 1/10 | >110 | |
| 10th–12th | 1/10 | >110 | |
| 13th–15th | 1/5 | >110 | |
| 16th–18th | 1/4 | >110 | |
| 19th–21st | 1/3 | >110 | |
| 22th–24th | 1/2 | >110 | |
| Yes | 1st–3rd | 0/6 | <0.3 |
| 4th–6th | 1/6 | 4.3 | |
| 7th–9th | 0/6 | <0.3 | |
| 10th–12th | 0/6 | <0.3 | |
| 13th–15th | 0/3 | <0.3 | |
| 16th–18th | 0/1 | <0.3 |
The present study was carried out to investigate possible routes of Salmonella contamination in leaf lettuce during cultivation. Once the soil is contaminated with pathogens such as Salmonella, there is the possibility that the bacteria persist in the soil throughout the cultivation period and contaminate the edible portions of vegetables by contact with contaminated soil. We first investigated Salmonella survival in soil during the 70-day cultivation period. As shown in Fig. 1, SE-EGFP counts increased once and were gradually reduced during the 70-day cultivation. Our previous study also showed that Salmonella survived in soil for >70 days during tomato cultivation (Mishima et al., 2012). The soil did not completely desiccate during cultivation, since irrigation was carried out daily. This might be one of the reasons for the survival of SE-EGFP in soil for 70 days. Islam et al. (2004) investigated the fate of Salmonella on carrots and radishes grown in fields treated with contaminated manure composts or irrigation water. Salmonella survived for 200 days even in soil, which had been inoculated once with Salmonella (105 cfu/g) suspended in irrigation water. Furthermore, Zhang et al. (2009) investigated the presence of E. coli O157:H7 on and in leaves of lettuce grown in inoculated soil. They showed that all samples from soil inoculated at 6 log cfu/g were positive for E. coli O157:H7 at 17 days after inoculation. Of 60 samples, 54 and 45 were positive for E. coli O157:H7 at 45 and 60 days after inoculation, respectively. With soil inoculation at 3 log cfu/g, all soil samples were positive for E. coli O157:H7 in the 26-day post inoculation samples, but only 16 of 60 samples were positive at 60 days. Our results and those of Zhang et al. (2009) indicated that all the soils were positive for Gram-negative foodborne pathogens such as Salmonella and E. coli O157:H7 for 60 to 70 days after soil contamination, even at low concentrations.
We investigated the relationship between spiking with Salmonella and contamination of lettuce (Tables 1 and 2). Contamination was not confirmed by investigating whole lettuce (Table 1). However, as shown in Table 2, Salmonella was detected by investigating sets of 3 outermost leaves. The Salmonella contamination level of the lettuce leaves at harvest increased with increasing initial inoculum levels of Salmonella in the soil. Not only the outer leaves but also the inner leaves of lettuce were contaminated with Salmonella when cultivated in soil highly contaminated with Salmonella. These results indicate that lettuce grown in soil containing Salmonella or E. coli O157:H7 may become contaminated throughout the growing season. Gagliardi and Karns (2002) revealed that E. coli O157:H7 survived for 25 to 41 days in fallow soils; >500 days in frozen soil; and 47 to 96 days on rye (Secale cereale L.), alfalfa (Medicago sativa), hairy vetch (Vicia sativa L.), and crimson clover (Trifolium incarnatum L.). In an E. coli O157:H7 survival study in Dutch soils, the calculated detection time by the Weibull model was 54 to 105 days. Franz et al. (2008) revealed that E. coli O157:H7 populations declined more rapidly under more oligotrophic soil conditions, which can be achieved by using manure with a relatively high carbon-to-nitrogen ratio and consequently, a relatively low rate of nutrient release. The pH and fiber content of manure used in soils also affected the survival of E. coli O157:H7 and Salmonella in soil (Franz et al., 2005). It seems that soil composition, temperature, and moisture content can influence pathogen survival, with nutritionally-rich soil in combination with moisture significantly increasing the soil persistence of E. coli O157:H7 (Vidovic et al., 2007). Considering that these factors can contribute to the contamination and persistence of foodborne pathogens, measures should be taken to prevent soil contamination with Gram-negative foodborne pathogens, such as Salmonella and E. coli O157:H7, from the time of seeding to harvesting to reduce the risk of contamination of lettuce at harvest.
In Tables 4 and 5, we investigated the possibility of internalization of Salmonella in lettuce leaves from contaminated soils by comparing Salmonella counts before and after surface disinfection. Salmonella counts of intact (non-injured) lettuces and lettuces with artificially-injured roots are shown in Tables 4 and 5, respectively. In both intact and root-injured lettuces, Salmonella was detected before disinfection; however, no Salmonella was detected after disinfection. Sodium hypochlorite solution is a sanitizer that is widely used for disinfecting vegetables and fruits. Furthermore, the solution is not thought to penetrate the interior of lettuce leaves because of its rigid structure, e.g., the cuticular layers. Thus, it has been used to investigate the internalization of bacteria into plants (Jablasone et al., 2005; Mitra et al., 2009). In the present study, by using ethanol and sodium hypochlorite solution, the surfaces of leaf lettuce were disinfected. After surface disinfection, Salmonella was not detected in most lettuce plants except for one (Table 4). Salmonella was detected from a set of the 1st - 3rd outermost leaves in one out of 27 disinfected plants, one of the corresponding leaves was etiolated and the injured part of the leaf was likely contaminated with Salmonella. Taken together, although Salmonella in soils contaminates the surfaces of lettuce leaves during cultivation, it appears that Salmonella is not internalized by intact lettuce plants. Zhang et al. (2009) reported that all 512 surface-disinfected leaf lettuce samples, regardless of plant age when inoculated or analyzed, type of lettuce, strain of E. coli O157:H7 in the inoculum, or water or cow manure extract used as an inoculum carrier, were negative for E. coli O157:H7. Taken together with our results, this indicates that the internalization of Gram-negative foodborne pathogens in lettuce tissues from the contaminated soil in which plants are grown does not occur.
During cultivation, water and soils, which are splashed by overhead irrigation and rainfall, might transfer contamination from the soil to the edible parts of vegetables. To analyze such contamination routes, the effect of mulching on contamination from spiked soils was studied. As shown in Tables 3 and 4, mulching reduced Salmonella contamination from soils to the edible parts of leaf lettuce as expected. Mulching likely reduced the opportunity for contact between the soil and the edible parts of lettuce plants. In the USA, overhead irrigation by sprinkler systems accounts for 58% of lettuce cultivation (Solomon et al., 2002). Overhead irrigation using such systems has the possibility of increasing the contamination levels of leafy vegetables with pathogens. Firstly, the weight of the irrigation water might allow contact between the lettuce leaves and the soil, leading to bacterial contact with the leaves. Secondly, the water splash containing soils and pathogenic bacteria might contact the leaves. In the present study, we employed mulch to prevent contact between contaminated soils and lettuce leaves. As shown in Table 6, overhead irrigation resulted in Salmonella contamination in only one lettuce plant. However, mulching certainly reduced contamination rates. The present study clearly showed that cultivation with mulch effectively prevented bacterial contamination of leafy vegetables.
Acknowledgements This work was supported by a grant from the Ministry of Agriculture, Forestry and Fisheries of Japan (Research project for ensuring food safety from farm to table FV-4103). The authors are grateful to Mr. Md. Tariqul Islam for the preparation of this manuscript.
