32 Journal o f Food Protection, Vol. 77, No. I, 2014, Pages 32-39 doi: 10.4315/0362-028X.JFP-13-019 Copyright © , International Association for Food Protection
In a c tiv a tio n o f S tr e s s e d
Escherichia coli 0 1 5 7 :H 7
C e lls o n th e
S u r fa c e s o f R o c k e t S a la d L e a v e s b y C h lo r in e a n d P e r o x y a c e tic A c id ANAS A. AL-NABULSI,'* TAREQ M. OSAILI,' HEBA M. OBAIDAT, 1 REYAD R. SHAKER , 1 SADDAM S. AWAISHEH, 2 RICHARD A. HOLLEY 3
and
1Department o f Nutrition and Food Technology, Jordan University o f Science and Technology, Irbid 22110, Jordan; departm ent o f Nutrition and Food Technology, Al-Balqa Applied University, Al-Salt 19117, Jordan; and 3Department o f Food Science, University o f Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 MS 13-019: Received 16 January 2013/Accepted 24 July 2013
A BSTR A CT Because Escherichia coli 0157:H7 has been frequently associated with many foodbome outbreaks caused by consumption of leafy greens (lettuce, spinach, and celery), this study investigated the ability of deionized water, chlorine, and peroxyacetic acid to detach or inactivate stressed and unstressed cells of E. coli 0157:H7 contaminating the surfaces of rocket salad leaves. E. coli 0157:H7 cells stressed by acid, cold, starvation, or NaCl exposure, as well as unstressed cells, were inoculated on the surfaces of rocket salad leaves at 4°C. The effectiveness of two sanitizers (200 ppm of chlorine and 80 ppm of peroxyacetic acid) and deionized water for decontaminating the leaves treated with stressed and unstressed E. coli 0157:H7 were evaluated during storage at 10 or 25°C for 0.5, 1, 3, and 7 days. It was found that washing with 80 ppm of peroxyacetic acid was more effective and reduced unstressed and stressed cells of E. coli 0157:H7 by about 1 log CFU per leaf on the leaves. There was no apparent difference in the ability of stressed and unstressed cells to survive surface disinfection with the tested agents. Treatments to reduce viable E. coli 0157:H7 cells on rocket leaves stored at 25°C were more effective than when used on those stored at 10°C. Washing with peroxyacetic acid or chlorine solution did not ensure the safety of rocket leaves, but such treatments could reduce the likelihood of water-mediated transfer of E. coli 0157:H7 during washing and subsequent processing.
The fresh produce industry has witnessed rapid growth due to market globalization in response to increased consumer demand for fresh, healthful, convenient, readyto-eat products. Improvements in packaging technology and distribution have been primarily responsible for the availability of different varieties of fresh produce and ready-to-eat salads (8). Almost in parallel, foodbome outbreaks associated with fresh produce consumption have increased dramatically (31). Pathogenic microorganisms may contaminate fresh produce at different points in the production chain at both pre- and postharvest levels. Preharvest contamination can occur from contaminated water used for irrigation, soil, or animal manure, whereas postharvest contamination may result from wash water, containers, processing equipment such as contaminated slicers and shredders, and improper handling by workers or consumers (18, 24). In addition, just a few contaminated leaves can serve to cross-contaminate a large mass of uncontaminated leaves (20, 41). Any of these circumstances can result in pathogen attachment to the surfaces of produce leaves, where the organisms can remain viable for long periods of time, depending on the storage temperature, available moisture, and leaf maturity (10, 15). * Author for correspondence. Tel: +962-02-7201000; Fax: +962-02-7201078; E-mail: [email protected].
The occurrence of pathogenic foodbome bacteria capable of causing human infections has been documented with virtually all types of fresh produce. Brackett (9) observed that enterotoxigenic and enterohemorrhagic Escherichia coli, Salmonella spp., Shigella spp., Campylobacter spp., Listeria monocytogenes, and Vibrio spp. have been isolated from many different types of raw fruit and vegetables. In October 2009 the Center for Science in the Public Interest in the United States noted that, based on the cases of illnesses reported between 1990 and 2007, leafy green vegetables occupied first place among the top ten riskiest foods regulated by the U.S. Food and Drug Administration (12). E. coli 0157:H7 has been frequently associated with many foodbome outbreaks caused by consumption of leafy greens, like lettuce, spinach, and celery. During 1995 to 2007, 23 outbreaks were associated with the consumption of green leafy vegetables contaminated with E. coli 0157:H7, and of these, 21 involved lettuce, whereas 2 were caused by spinach (12). Early in the summer of 2011, a major outbreak with nearly 4,000 illnesses resulted from the consumption of sprouted fenugreek seeds contaminated with E. coli O104:H4 in northern Germany. This outbreak caused illnesses among visitors from 15 countries in Europe and North America (17). E. coli 0157:H7 can survive under adverse conditions in the soil for long periods of time. Mukherjee et al. (28)
J. Food Prot., Vol. 77, No. 1
INACTIVATION OF STRESSED E. COLl 0157:H7 ON ROCKET LEAF SURFACES
reported that E. coli 0157:H7 has the capability to survive for 4 to 8 weeks in the soil, but others have found survival up to 25 weeks (31). The survival of E. coli 0157:H7 is improved at lower temperatures, in clay soil, and in close association with roots (21,22). In response to environmental stress, foodborne pathogens may become better suited to survive and possibly grow during subsequent processing or may become more virulent (36). Furthermore, the folds, crevices, and pores of leafy green vegetables may protect foodborne pathogens and make it difficult for them to be detached or inactivated by conventional sanitizers like chlorine (12). In an attempt to reduce the risk associated with the contamination of green leafy vegetables with E. coli 0157:H7 and other foodborne pathogens, different strate gies have been investigated, including the use of disinfec tants; however, each has limitations. Chlorine, widely used for the disinfection of leafy green vegetables, is inactivated when in contact with organic materials, and it can form compounds with negative effects on human health (40). This has highlighted the need for alternatives; and disinfectants such as chlorine dioxide in the gaseous or liquid forms, peroxyacetic acid, acidic electrolyzed water, and other approaches have been proposed (27, 31). Rocket leaves are used as an ingredient in green salads or are consumed alone as a side dish. Although the outbreaks associated with consumption of rocket leaves are less frequent compared to other fresh produce, Nygard et al. (29) reported that rocket leaves contaminated with Salmonella enterica were involved in an outbreak that took place in the United Kingdom and Scandinavia in 2004. Furthermore, Russell et al. (34) reported that 100% of the rocket leaves sampled were contaminated with E. coli at a level of ~ 6 log CFU/g. Previous reports showed that rocket leaves were colonized by enterotoxigenic E. coli and enteroaggregative E. coli, using flagella and aggregative adherence fimbriae, respectively (5, 31, 38). It is worth noting that, in these studies, E. coli 0157:H7 cells were grown under optimal laboratory conditions; however, bacterial cells may encounter substantial environmental stress during transit from the intestines of infected animals on the farm and through produce distribution before they reach consumers (15). The objective of the current study was to investigate the ability of deionized water, chlorine, and peroxyacetic acid to detach or inactivate stressed and unstressed cells of E. coli 0157:H7 at the surfaces of rocket salad leaves. MATERIALS AND METHODS Preparation of rocket leaves. Rocket leaves were purchased from a local supermarket in Irbid, North Jordan, on the day o f each experiment. Damaged leaves were removed; undamaged leaves were washed with sterilized water, dried using a salad spinner, and then screened for the presence o f E. coli 0157:H 7 before the beginning of each experiment. Inoculum preparation. Four clinical isolates o f E. coli 0157:H 7 (00:0304, 02:0627, 02:0628, and 02:3581) that had become nonpathogenic (verotoxigenic negative) during storage
33
were provided by Rafiq Ahmed, National Microbiology Labora tory, Public Health Agency, Canadian Science Centre for Human and Animal Health, Winnipeg, Manitoba, Canada. All strains were stored individually at —40°C in tryptic soy broth (TSB; Oxoid Ltd., Basingstoke, UK) containing 20% (vol/vol) glycerol (SigmaAldrich, St. Louis, MO). To activate the microbes for experimental use, a loopful from each culture was inoculated into 10 ml o f sterile TSB and incubated at 37°C for 24 h. The cultures then were stored at 4°C in sorbitol MacConkey agar. Prior to the experiment, the cultures were individually inoculated into 10 ml of sterile TSB and incubated at 37°C for 24 h. A mixture containing equal volumes of each o f the four strains was prepared by mixing 2.5 ml of each culture. The cocktail mixture was centrifuged at 4,000 x g for 20 min, the supernatant was discarded, and the pellet was washed with sterile deionized water. The pellet was resuspended in 10 ml o f sterile deionized water, and the suspension was diluted in sterile deionized water to achieve 108 CFU/ml (30).
Preparation of acid-stressed E. coli 0157:H7. To prepare acid-adapted cells, a loopful o f each culture was transferred into 10 ml of TSB supplemented with 10 g/liter glucose and was incubated for 18 h at 37°C to reach 109 CFU/ml. The final pH following incubation was 4.9 ± 0.1 (26). Then a mixture was prepared containing equal volumes o f each strain as described above. Preparation of starvation-stressed E. coli 0157:H7. To prepare starved cells, a loopful of each culture was inoculated into 10 ml of TSB and incubated for 18 h to reach 109 CFU/ml. A cocktail was prepared containing equal volumes of each strain, as described above, in saline solution (0.85% NaCl, pH 6.6) (26) and was incubated for 48 h at 37°C. Preparation of cold-stressed E. coli 0157:H7. Coldstressed cells o f E. coli 0157:H 7 were prepared by inoculating a loopful o f each culture into 10 ml o f TSB for 18 h (109 CFU/ml). A culture cocktail, prepared containing equal volumes o f each strain as described above, was resuspended in 10 ml of TSB and incubated for 7 days at 5°C (26). Preparation of salt-stressed E. coli 0157:H7. Salt-stressed E. coli 0157:H 7 cells were prepared by inoculating a loopful of each culture into 10 ml o f TSB supplemented with NaCl (3.5%) for 18 h (109 CFU/ml). A cocktail was prepared containing equal volumes of each strain as described above and was resuspended in 10 m l of sterile deionized water (19).
Inoculation of rocket leaf surfaces by the spot method. The surfaces of washed rocket leaves were inoculated by adding 50 pi o f stressed and unstressed E. coli 0157:H 7 cells at five different locations per leaf (on the upper surface). The inoculated leaves were dried in a biosafety cabinet for 2 h at 25°C, and the leaves were kept at 4°C for 22 h to allow E. coli 0157:H 7 to adhere to the surfaces o f the inoculated leaves, to simulate pre- or postharvest contamination, at least 24 h before leaves were washed with different sanitizers (7, 25). Test samples were stored at 25 or 10°C for up to 7 days.
Preparation of sanitizer solutions. Chlorine solution was freshly prepared before each experiment by diluting sodium hypochlorite (Scharlau Chemie S.A., Barcelona, Spain) with 0.05 M potassium phosphate buffer (pH 6.8 at 25°C) to 200 ppm (25). Peroxyacetic acid solution was freshly prepared for each experiment by diluting peroxyacetic acid (Sigma-Aldrich) in sterile distilled water to obtain 80 ppm.
J. Food Prot., Vol. 77, No. 1
AL-NABULSI ET AL.
Sanitizing treatm ent. After 0.5, 1, 3, and 7 days of storage at 10 or 25°C, the inoculated rocket leaf (—0.85 g) was placed into a stomacher bag containing 100 ml of each test solution (deionized water, 80 ppm of peroxyacetic acid, or 200 ppm of chlorine solution). The bag containing rocket leaves and sanitizer solution was agitated vigorously by hand for 5 min to simulate a commercial wash operation. The leaves were transferred to a salad spinner to remove excess liquid from leaf surfaces (23,33), and the leaves were analyzed for the presence of E. coli 0157:H7. Results were expressed as numbers of bacteria per leaf. Microbiological analysis. Each rocket leaf was transferred into a sterile stomacher bag containing 100 ml of 0.1 N sodium thiosulphate to neutralize the residual sanitizing solution and was serially diluted in 0.1% peptone water (33). From appropriate dilutions, 0.1 ml was plated on the surface of sorbitol MacConkey agar supplemented with 0.05 mg/liter cefixime and 2.5 mg/liter potassium tellurite and overlaid with tryptic soy agar to facilitate the growth of injured cells. The plates were then incubated at 37°C for 18 to 24 h (23). Statistical analysis. Statistical analysis was done with SPSS software, version 19 (IBM Inc., Armonk, NY) using a univariate general linear model, in which the mean and standard deviation were calculated for each variable. Each experiment was replicated three times, and each replicate consisted of two rocket leaves.
u
CQ
’7}-
< cu
d
u X cd
cd
u X
CQ
X
cd
»n OO ON CM OC o o o
o
+i +i +i +i +i CM in in in
u
n o o
CQ
U
O X
X cd
u O
ON CM
NO
ON co
o
d
o
+i +i +i +i +i CM CM c» CO ON in o o 0 0 of; in in "d- >n
§
CQ CQ
< X
cd
cd
u
SC o hJ
u
CQ
< X
< o
in O n O n ON r- CO CM — o o d d +i -h
< CJ »n in o
+i +i +i
>n T7f o O n TT in in in »n CQ
3 ^
CQ CQ < cd cd CO CN 0 0 CN CO CN o d o o o < cd
£ Q
CQ < cd
CQ < cd
n CO o r80
In a c tiv a tio n o f S tr e s s e d
Escherichia coli 0 1 5 7 :H 7
C e lls o n th e
S u r fa c e s o f R o c k e t S a la d L e a v e s b y C h lo r in e a n d P e r o x y a c e tic A c id ANAS A. AL-NABULSI,'* TAREQ M. OSAILI,' HEBA M. OBAIDAT, 1 REYAD R. SHAKER , 1 SADDAM S. AWAISHEH, 2 RICHARD A. HOLLEY 3
and
1Department o f Nutrition and Food Technology, Jordan University o f Science and Technology, Irbid 22110, Jordan; departm ent o f Nutrition and Food Technology, Al-Balqa Applied University, Al-Salt 19117, Jordan; and 3Department o f Food Science, University o f Manitoba, Winnipeg, Manitoba, Canada R3T 2N2 MS 13-019: Received 16 January 2013/Accepted 24 July 2013
A BSTR A CT Because Escherichia coli 0157:H7 has been frequently associated with many foodbome outbreaks caused by consumption of leafy greens (lettuce, spinach, and celery), this study investigated the ability of deionized water, chlorine, and peroxyacetic acid to detach or inactivate stressed and unstressed cells of E. coli 0157:H7 contaminating the surfaces of rocket salad leaves. E. coli 0157:H7 cells stressed by acid, cold, starvation, or NaCl exposure, as well as unstressed cells, were inoculated on the surfaces of rocket salad leaves at 4°C. The effectiveness of two sanitizers (200 ppm of chlorine and 80 ppm of peroxyacetic acid) and deionized water for decontaminating the leaves treated with stressed and unstressed E. coli 0157:H7 were evaluated during storage at 10 or 25°C for 0.5, 1, 3, and 7 days. It was found that washing with 80 ppm of peroxyacetic acid was more effective and reduced unstressed and stressed cells of E. coli 0157:H7 by about 1 log CFU per leaf on the leaves. There was no apparent difference in the ability of stressed and unstressed cells to survive surface disinfection with the tested agents. Treatments to reduce viable E. coli 0157:H7 cells on rocket leaves stored at 25°C were more effective than when used on those stored at 10°C. Washing with peroxyacetic acid or chlorine solution did not ensure the safety of rocket leaves, but such treatments could reduce the likelihood of water-mediated transfer of E. coli 0157:H7 during washing and subsequent processing.
The fresh produce industry has witnessed rapid growth due to market globalization in response to increased consumer demand for fresh, healthful, convenient, readyto-eat products. Improvements in packaging technology and distribution have been primarily responsible for the availability of different varieties of fresh produce and ready-to-eat salads (8). Almost in parallel, foodbome outbreaks associated with fresh produce consumption have increased dramatically (31). Pathogenic microorganisms may contaminate fresh produce at different points in the production chain at both pre- and postharvest levels. Preharvest contamination can occur from contaminated water used for irrigation, soil, or animal manure, whereas postharvest contamination may result from wash water, containers, processing equipment such as contaminated slicers and shredders, and improper handling by workers or consumers (18, 24). In addition, just a few contaminated leaves can serve to cross-contaminate a large mass of uncontaminated leaves (20, 41). Any of these circumstances can result in pathogen attachment to the surfaces of produce leaves, where the organisms can remain viable for long periods of time, depending on the storage temperature, available moisture, and leaf maturity (10, 15). * Author for correspondence. Tel: +962-02-7201000; Fax: +962-02-7201078; E-mail: [email protected].
The occurrence of pathogenic foodbome bacteria capable of causing human infections has been documented with virtually all types of fresh produce. Brackett (9) observed that enterotoxigenic and enterohemorrhagic Escherichia coli, Salmonella spp., Shigella spp., Campylobacter spp., Listeria monocytogenes, and Vibrio spp. have been isolated from many different types of raw fruit and vegetables. In October 2009 the Center for Science in the Public Interest in the United States noted that, based on the cases of illnesses reported between 1990 and 2007, leafy green vegetables occupied first place among the top ten riskiest foods regulated by the U.S. Food and Drug Administration (12). E. coli 0157:H7 has been frequently associated with many foodbome outbreaks caused by consumption of leafy greens, like lettuce, spinach, and celery. During 1995 to 2007, 23 outbreaks were associated with the consumption of green leafy vegetables contaminated with E. coli 0157:H7, and of these, 21 involved lettuce, whereas 2 were caused by spinach (12). Early in the summer of 2011, a major outbreak with nearly 4,000 illnesses resulted from the consumption of sprouted fenugreek seeds contaminated with E. coli O104:H4 in northern Germany. This outbreak caused illnesses among visitors from 15 countries in Europe and North America (17). E. coli 0157:H7 can survive under adverse conditions in the soil for long periods of time. Mukherjee et al. (28)
J. Food Prot., Vol. 77, No. 1
INACTIVATION OF STRESSED E. COLl 0157:H7 ON ROCKET LEAF SURFACES
reported that E. coli 0157:H7 has the capability to survive for 4 to 8 weeks in the soil, but others have found survival up to 25 weeks (31). The survival of E. coli 0157:H7 is improved at lower temperatures, in clay soil, and in close association with roots (21,22). In response to environmental stress, foodborne pathogens may become better suited to survive and possibly grow during subsequent processing or may become more virulent (36). Furthermore, the folds, crevices, and pores of leafy green vegetables may protect foodborne pathogens and make it difficult for them to be detached or inactivated by conventional sanitizers like chlorine (12). In an attempt to reduce the risk associated with the contamination of green leafy vegetables with E. coli 0157:H7 and other foodborne pathogens, different strate gies have been investigated, including the use of disinfec tants; however, each has limitations. Chlorine, widely used for the disinfection of leafy green vegetables, is inactivated when in contact with organic materials, and it can form compounds with negative effects on human health (40). This has highlighted the need for alternatives; and disinfectants such as chlorine dioxide in the gaseous or liquid forms, peroxyacetic acid, acidic electrolyzed water, and other approaches have been proposed (27, 31). Rocket leaves are used as an ingredient in green salads or are consumed alone as a side dish. Although the outbreaks associated with consumption of rocket leaves are less frequent compared to other fresh produce, Nygard et al. (29) reported that rocket leaves contaminated with Salmonella enterica were involved in an outbreak that took place in the United Kingdom and Scandinavia in 2004. Furthermore, Russell et al. (34) reported that 100% of the rocket leaves sampled were contaminated with E. coli at a level of ~ 6 log CFU/g. Previous reports showed that rocket leaves were colonized by enterotoxigenic E. coli and enteroaggregative E. coli, using flagella and aggregative adherence fimbriae, respectively (5, 31, 38). It is worth noting that, in these studies, E. coli 0157:H7 cells were grown under optimal laboratory conditions; however, bacterial cells may encounter substantial environmental stress during transit from the intestines of infected animals on the farm and through produce distribution before they reach consumers (15). The objective of the current study was to investigate the ability of deionized water, chlorine, and peroxyacetic acid to detach or inactivate stressed and unstressed cells of E. coli 0157:H7 at the surfaces of rocket salad leaves. MATERIALS AND METHODS Preparation of rocket leaves. Rocket leaves were purchased from a local supermarket in Irbid, North Jordan, on the day o f each experiment. Damaged leaves were removed; undamaged leaves were washed with sterilized water, dried using a salad spinner, and then screened for the presence o f E. coli 0157:H 7 before the beginning of each experiment. Inoculum preparation. Four clinical isolates o f E. coli 0157:H 7 (00:0304, 02:0627, 02:0628, and 02:3581) that had become nonpathogenic (verotoxigenic negative) during storage
33
were provided by Rafiq Ahmed, National Microbiology Labora tory, Public Health Agency, Canadian Science Centre for Human and Animal Health, Winnipeg, Manitoba, Canada. All strains were stored individually at —40°C in tryptic soy broth (TSB; Oxoid Ltd., Basingstoke, UK) containing 20% (vol/vol) glycerol (SigmaAldrich, St. Louis, MO). To activate the microbes for experimental use, a loopful from each culture was inoculated into 10 ml o f sterile TSB and incubated at 37°C for 24 h. The cultures then were stored at 4°C in sorbitol MacConkey agar. Prior to the experiment, the cultures were individually inoculated into 10 ml of sterile TSB and incubated at 37°C for 24 h. A mixture containing equal volumes of each o f the four strains was prepared by mixing 2.5 ml of each culture. The cocktail mixture was centrifuged at 4,000 x g for 20 min, the supernatant was discarded, and the pellet was washed with sterile deionized water. The pellet was resuspended in 10 ml o f sterile deionized water, and the suspension was diluted in sterile deionized water to achieve 108 CFU/ml (30).
Preparation of acid-stressed E. coli 0157:H7. To prepare acid-adapted cells, a loopful o f each culture was transferred into 10 ml of TSB supplemented with 10 g/liter glucose and was incubated for 18 h at 37°C to reach 109 CFU/ml. The final pH following incubation was 4.9 ± 0.1 (26). Then a mixture was prepared containing equal volumes o f each strain as described above. Preparation of starvation-stressed E. coli 0157:H7. To prepare starved cells, a loopful of each culture was inoculated into 10 ml of TSB and incubated for 18 h to reach 109 CFU/ml. A cocktail was prepared containing equal volumes of each strain, as described above, in saline solution (0.85% NaCl, pH 6.6) (26) and was incubated for 48 h at 37°C. Preparation of cold-stressed E. coli 0157:H7. Coldstressed cells o f E. coli 0157:H 7 were prepared by inoculating a loopful o f each culture into 10 ml o f TSB for 18 h (109 CFU/ml). A culture cocktail, prepared containing equal volumes o f each strain as described above, was resuspended in 10 ml of TSB and incubated for 7 days at 5°C (26). Preparation of salt-stressed E. coli 0157:H7. Salt-stressed E. coli 0157:H 7 cells were prepared by inoculating a loopful of each culture into 10 ml o f TSB supplemented with NaCl (3.5%) for 18 h (109 CFU/ml). A cocktail was prepared containing equal volumes of each strain as described above and was resuspended in 10 m l of sterile deionized water (19).
Inoculation of rocket leaf surfaces by the spot method. The surfaces of washed rocket leaves were inoculated by adding 50 pi o f stressed and unstressed E. coli 0157:H 7 cells at five different locations per leaf (on the upper surface). The inoculated leaves were dried in a biosafety cabinet for 2 h at 25°C, and the leaves were kept at 4°C for 22 h to allow E. coli 0157:H 7 to adhere to the surfaces o f the inoculated leaves, to simulate pre- or postharvest contamination, at least 24 h before leaves were washed with different sanitizers (7, 25). Test samples were stored at 25 or 10°C for up to 7 days.
Preparation of sanitizer solutions. Chlorine solution was freshly prepared before each experiment by diluting sodium hypochlorite (Scharlau Chemie S.A., Barcelona, Spain) with 0.05 M potassium phosphate buffer (pH 6.8 at 25°C) to 200 ppm (25). Peroxyacetic acid solution was freshly prepared for each experiment by diluting peroxyacetic acid (Sigma-Aldrich) in sterile distilled water to obtain 80 ppm.
J. Food Prot., Vol. 77, No. 1
AL-NABULSI ET AL.
Sanitizing treatm ent. After 0.5, 1, 3, and 7 days of storage at 10 or 25°C, the inoculated rocket leaf (—0.85 g) was placed into a stomacher bag containing 100 ml of each test solution (deionized water, 80 ppm of peroxyacetic acid, or 200 ppm of chlorine solution). The bag containing rocket leaves and sanitizer solution was agitated vigorously by hand for 5 min to simulate a commercial wash operation. The leaves were transferred to a salad spinner to remove excess liquid from leaf surfaces (23,33), and the leaves were analyzed for the presence of E. coli 0157:H7. Results were expressed as numbers of bacteria per leaf. Microbiological analysis. Each rocket leaf was transferred into a sterile stomacher bag containing 100 ml of 0.1 N sodium thiosulphate to neutralize the residual sanitizing solution and was serially diluted in 0.1% peptone water (33). From appropriate dilutions, 0.1 ml was plated on the surface of sorbitol MacConkey agar supplemented with 0.05 mg/liter cefixime and 2.5 mg/liter potassium tellurite and overlaid with tryptic soy agar to facilitate the growth of injured cells. The plates were then incubated at 37°C for 18 to 24 h (23). Statistical analysis. Statistical analysis was done with SPSS software, version 19 (IBM Inc., Armonk, NY) using a univariate general linear model, in which the mean and standard deviation were calculated for each variable. Each experiment was replicated three times, and each replicate consisted of two rocket leaves.
u
CQ
’7}-
< cu
d
u X cd
cd
u X
CQ
X
cd
»n OO ON CM OC o o o
o
+i +i +i +i +i CM in in in
u
n o o
CQ
U
O X
X cd
u O
ON CM
NO
ON co
o
d
o
+i +i +i +i +i CM CM c» CO ON in o o 0 0 of; in in "d- >n
§
CQ CQ
< X
cd
cd
u
SC o hJ
u
CQ
< X
< o
in O n O n ON r- CO CM — o o d d +i -h
< CJ »n in o
+i +i +i
>n T7f o O n TT in in in »n CQ
3 ^
CQ CQ < cd cd CO CN 0 0 CN CO CN o d o o o < cd
£ Q
CQ < cd
CQ < cd
n CO o r80
