Fates of Seeded Escherichia coli O157:H7 and Salmonella on Selected Fresh Culinary Herbs during Refrigerated Storage

Transcription

1 1997 Journal of Food Protection, Vol. 69, No. 8, 2006, Pages Copyright, International Association for Food Protection Research Note Fates of Seeded Escherichia coli O157:H7 and on Selected Fresh Culinary Herbs during Refrigerated Storage WEI-YEA HSU, 1 AMARAT SIMONNE, 1 * A PONGPHEN JITAREERAT 2 1 Department of Family, Youth and Community Sciences, 3028 McCarty Hall, P.O. Box , University of Florida, Gainesville, Florida , USA; and 2 Division of Postharvest Technology, School of Bioresources and Technology, King Mongkut s University of Technology Thonburi, Thung-Kru, Bang-Mod, Bangkok, Thailand MS : Received 30 November 2005/Accepted 15 March 2006 ABSTRACT The fates of seeded Escherichia coli O157:H7 and on selected fresh culinary herbs were evaluated at a refrigerated temperature (4 C). Fresh herbs, including cilantro, oregano, basil, chive, parsley, and rosemary, were inoculated with six-strain mixtures of and, and the microbial populations were monitored at 1, 5, 11, 16, 19, and 24 days. For both pathogens, a significant decrease in the population (P ) occurred within the first 5 days of storage ( 0.8 log). Both pathogens remained the highest on cilantro and the lowest on rosemary (P ). Storage time had a significant effect on the survival of ; populations declined as storage time progressed. Although storage of cilantro, basil, and chive was terminated after 19 days because of deteriorated quality, significant numbers of both pathogens were recovered from the remaining fresh herbs after 24 days of storage. The results showed that both bacteria were extremely persistent on all test herbs under the test conditions. The results also reinforce the concept that, once contaminated, bacterial pathogens can persist on fresh herbs throughout a normal distribution time. The numbers of foodborne outbreaks traced back to fresh fruits and vegetables have increased in recent years (4 6, 10, 16, 17, 20, 22, 23), and many of those outbreaks are associated with contaminated fresh culinary herbs (4, 5, 10, 16, 22, 23). Fresh culinary herbs are leafy plant materials used in small amounts to add aroma and flavor to foods. Because of their unique flavor profile, fresh herbs from various origins are used in foods as a garnish, often without cooking. Since fresh herbs are often consumed raw without further processing to inactivate harmful microorganisms, they can pose a potential risk of foodborne illness for consumers. For example, in 1998, chopped, uncooked curly parsley was implicated in the outbreaks of shigellosis in four U.S. states and two Canadian provinces; all fresh parsley consumed in all the outbreak cases was contaminated with Shigella sonnei (5, 24). In 1999, fresh cilantro was implicated in an outbreak of salmonellosis; the herb was contaminated with Thompson (2, 4). Furthermore, several outbreaks of cyclosporiasis associated with fresh basil were reported in British Columbia, Canada, in 2001 (10) and in Missouri in 1999 (16). While the vehicle of the cyclosporiasis infection in Missouri was chicken pasta salads garnished with fresh basil (16), the sporadic outbreak of cyclosporiasis in Canada was linked to the consumption of imported Thai basil, which is used as a part of Vietnamese foods (10); the basil was served raw as a garnish. * Author for correspondence. Tel: , Ext 232; Fax: ; asim@ufl.edu. Because of the increase in numbers of outbreaks of foodborne illnesses in fresh fruits and vegetables, in 1999 and 2000, the U.S. Food and Drug Administration (FDA) initiated a massive survey of high-volume imported (n 1,003) and domestic (n 1,000) fresh produce (22, 23) for the presence of and Escherichia coli O157:H7 or Shigella. Four percent of imported samples were contaminated with either Shigella or, and cilantro and culantro accounted for 1.6 and 0.6%, respectively. Eighty percent of the contaminated samples had (22). For domestic fresh produce tested, 1.1% was found positive for or Shigella; cilantro, parsley, and scallions had incidence rates of 0.1, 0.1, and 0.3%, respectively (23). Even though was not found on any of the FDA-surveyed produce, the potential risk of this pathogen on fresh herbs exists. Outbreaks of E. coli O157: H7 infections from fresh produce have occurred with increasing frequency in recent years (1, 9, 11, 15, 17, 20). Sources of and other pathogen contamination on fresh leafy produce could come from manure and irrigation water used in the fields (6, 11, 12, 21). A study demonstrated that could persist in soils treated with contaminated compost or irrigation water for more than 5 months, and it was detected on parsley grown on such soil for up to 177 days after seedlings were planted (11). Although the prevalence and survival of pathogenic bacteria, e.g., Campylobacter jejuni (14), (1, 9, 11, 15), Helicobacter pylori (19), Listeria monocy-

2 1998 HSU ET AL. J. Food Prot., Vol. 69, No. 8 togenes (15), and (12, 15), on various fresh produce are well documented, very little research has been done on the survival of different pathogenic bacteria on herbs. Survival and growth characteristics of inoculated S. sonnei on whole and chopped parsley leaves held at 4 and 21 C were studied (24); results indicated that S. sonnei grew rapidly on chopped parsley held at 21 C (ambient temperature) and remained viable for at least 14 days at 4 C. These particular findings support the results of the outbreak investigation of the S. sonnei associated with parsley in the United States and Canada in 1998; in that outbreak, chopped parsley held at room temperature was used as an ingredient for multiple dishes in each of the outbreak-associated restaurants (18). In parallel with the increase in the global distribution and consumption of fresh herbs, the number of multistate, multinational foodborne illness outbreaks associated with imported or domestic fresh herbs is on the rise. Extensive research is needed to fully understand the role of fresh herbs in the transmission of bacterial pathogens. Therefore, the objective of this research was to investigate the survival and growth characteristics of and on fresh herbs (chive, cilantro, parsley, basil, oregano, and rosemary) that are commonly used and consumed raw in many dishes. First, the population of and on test herbs was studied over 24 days of storage at 4 C. From these results, effects of herb type and storage time on the survival of both pathogens were evaluated. MATERIALS A METHODS Fresh herbs. Fresh herbs, including cilantro (Coriandrum sativum), oregano (Origanum vulgare), basil (Ocimum basilicum), chive (Allium schoenoprasum), parsley (Petroselinum crispum), and rosemary (Rosmarinus officinalis), were purchased from a commercial herb farm in Live Oak, Fla. The herbs were harvested from the greenhouse in the early morning to minimize field heat and packed in polyethylene terephthalate clamshell boxes (8 by 15.2 by 2.5 cm; 25 2 g per box). The temperature in the packing facility was 4 C. The herbs did not receive washing or sanitizing treatments. Samples were then transported in insulated boxes to the research laboratory on the University of Florida campus at Gainesville. The estimated transportation time was 45 min. In the afternoon of the same day, the herbs were inoculated with test bacterial cultures, subjected to storage at 4 C, and kept at 4 C throughout the experimentation period. Preparation of bacterial cultures. Three nalidixic acid resistant strains of (204P, 301C, and 505B) and three nalidixic acid resistant strains of Enteritidis, Typhimurium, and Mission obtained from Dr. Tun-Shi Huang (Auburn University, Auburn, Ala.) were utilized as inocula. One day prior to experimentation, nalidixic acid resistant bacterial cultures were streaked on tryptic soy agar (Difco, Becton Dickinson, Sparks, Md.) supplemented with 50 g/ml of nalidixic acid and incubated at 37 C overnight. One isolated colony of each overnight culture was transferred to an individual flask containing 200 ml of tryptic soy broth (Difco, Becton Dickinson) supplemented with 50 g/ml of nalidixic acid and incubated overnight at 37 C. After overnight incubation, each bacterial culture was harvested individually by centrifugation at 2,000 g for 15 min and washed once in 0.1% peptone water (Difco, Becton Dickinson). The final cell pellet was resuspended in 0.1% peptone water to achieve a final cell density of approximately CFU/ml. Equal volumes of each bacterial suspension were combined, and the mixtures were used as cocktails for herb inoculation. Herb inoculation. The spot inoculation technique (15) was employed to inoculate all herb samples. Aliquots (150 l) of cocktails, prepared as described, were applied to surfaces of herbs in each clamshell box in small volumes at 35 to 40 locations to facilitate drying. After the inoculation, samples were placed to dry in two-layer stackable shelves elevated 10 cm above the work surface in a laminar flow biosafety cabinet for 20 min. Herb storage. Once dried, samples in clamshells were stored in a cold room at 4 C, where herb growers packed and stored fresh-cut herbs before they were shipped to retailers. The clamshells were packed (approximately six clamshells per box) in a rectangular commercial cardboard box (10 by 19 by 22 cm). Herb quality was evaluated subjectively on the basis of the visual rating as described by Kader (13) throughout the test period. The storage study was terminated when the quality of samples had deteriorated to poor with major defects. Deterioration of appearance was judged as a loss of typical color, development of sliminess and browning, and weeping of tissue fluid. Microbiological enumeration and analysis. Pathogen populations (log CFU per gram) on the herb samples were evaluated over the storage period at 1, 5, 11, 16, 19, and 24 days. The inoculated herb in each clamshell box was processed as one sample. The content of each clamshell was placed in a 400-ml stomacher bag (Seward Scientific, London, UK) containing 200 ml of 0.1% peptone water and pummeled in a stomacher (Seward) for 1 min at normal speed. The aliquot was used for analyses of bacterial populations. Both undiluted and serially diluted aliquots (0.1 ml in duplicate) in 0.1% peptone water were surface plated on sorbitol MacConkey agar (Difco, Becton Dickinson) and bismuth sulfite agar (Difco, Becton Dickinson) containing 50 g/ml of nalidixic acid and 0.1% sodium pyruvate for enumeration of E. coli O157:H7 and, respectively. Sodium pyruvate was added to both direct plating media to enhance the recovery of injured cells (7, 15). After plates were incubated at 37 C for 24 h, presumptive colonies (five for each sample at each time point per experiment) on differential media were randomly selected and confirmed by agglutination tests. Colonies were confirmed as and by the latex agglutination test (Oxoid, Basingstoke, UK) and the Prolex E. coli O157 test kit (Fisher Scientific Laboratories, Pittsburgh, Pa.), respectively. Statistical analysis. The experiment was replicated two times and each time in triplicate for each herb at each sampling point, with a total of 90 samples for each time (six herbs, five time points, and three samples for each time point). Results of each herb type from all time points were used to evaluate the relationship of herbs and the survival of and. For the overall storage effect, results of all six herbs at each time point were used. Data (log CFU per gram) were analyzed by the general linear model procedure (PROC GLM; SAS System for Window 9, SAS Institute, Cary, N.C.). Mean separation was achieved by Duncan s multiple range test at RESULTS A DISCUSSION Changes in and populations on the inoculated herbs were monitored for 19 to 24

3 J. Food Prot., Vol. 69, No. 8 SURVIVAL OF E. COLI O157:H7 A SALMONELLA ON HERBS 1999 TABLE 1. Survival of seeded and on herbs during storage at 4 C Population (log CFU/g [fresh wt]) a Storage (day) Basil Chive Cilantro Oregano Parsley Rosemary A 5.01 BC 4.94 BC 4.72 BC b 5.44 A 5.24 A 4.98 AB 4.66 BC 4.43 C 5.65 A 5.68 A 5.25 B 4.99 C 4.77 D 5.50 A 5.38 A 5.01 B 4.76 B 4.75 B 4.71 B 5.53 A 5.02 B 4.76 C 4.56 CD 4.39 D 4.37 D 4.96 A 4.46 B 4.14 B 3.68 C 3.53 C 3.34 C A 5.13 B 5.07 B 4.91 B 5.41 A 5.09 AB 5.25 AB 4.96 BC 4.71 C 6.06 A 5.59 B 5.60 B 5.44 BC 5.24 C 5.83 A 5.28 B 5.21 BC 4.90 CD 4.89 CD 4.68 D 5.76 A 5.14 B 5.10 B 4.72 C 4.69 C 4.52 C 5.42 A 4.63 B 4.67 B 4.37 BC 4.31 BC 4.08 C a Means reflect the average of two experiments (total of six samples). For each bacterium, means in a column followed by different letters are significantly different by Duncan s multiple range test ( 0.05). b, not determined. days of storage at 4 C. The bacterial population of test herbs was evaluated at different time points throughout the study. However, when the sample quality was rated poor with major defects as previously described, the microbial study was terminated. The appearances of basil, chive, and cilantro were poor after 19 days of storage; thus, the microbial populations on these herbs were not determined thereafter. However, for oregano, parsley, and rosemary, the visual quality remained fair after 24 days of storage. A significant decrease ( 0.51 log CFU/g, P ) in populations was observed on basil, parsley, and rosemary within the first 5 days of storage (Table 1). For cilantro and oregano, a significant decrease did not occur until 11 days of storage. Populations of on cilantro, oregano, basil, parsley, and rosemary significantly decreased (0.47 to 0.8 log CFU/g) within the first 5 days of storage (Table 1). The bacterial population continued to decline after day 5 but was not significant until after either 16 or 19 days of storage. Although numbers of both test bacterial populations on chive decreased as storage time progressed, the change was not significant until 16 days of storage (Table 1). The highest reduction of E. coli O157:H7 and populations (1.62 and 1.34 log CFU/g, respectively) over 24 days of storage occurred on rosemary (Table 1). According to Wu et al. (24), populations of S. sonnei on inoculated chopped or whole parsley leaves kept at 4 C were reduced by 2.5 to 3.0 log CFU/g during a 14-day storage period. In our study, both and on parsley were reduced by only approximately 1.0 log CFU/g within 16 days of storage, the average shelf life of fresh produce (Table 1). Besides bacterial variation, the use of polyethylene terephthalate clamshell boxes for sample storage could be the cause of the lower reduction in populations. It was noted that weight loss of samples was minimized when they were kept tight in clamshell boxes stored at 4 C (data not shown). Test herbs used in this study were packed in clamshells and stacked in cardboard boxes throughout the entire storage period; this setting is one of the currently used situations in commercial distribution and storage of fresh herbs before displaying in retailer stores. Results demonstrate that if herbs are contaminated with these bacteria in this situation, they could survive on the surface until reaching the consumer. The typical shelf life of herbs throughout the commercial distribution is 2 weeks. The effect of herb type on the survival of both test pathogens was examined (Table 2). Overall, cilantro had the highest populations of and (5.27 and 5.59 log CFU/g, respectively). In contrast to cilantro, rosemary had the lowest populations of E. coli TABLE 2. Relationship between herb type and the survival of and stored at 4 C Population (log CFU/g [fresh wt]) a Pathogen Basil Chive Cilantro Oregano Parsley Rosemary 5.08 B 5.18 B 4.95 B 5.08 BC 5.27 A 5.59 A 5.02 B 5.13 BC 4.77 C 4.99 C 4.02 D 4.58 D a Means reflect the average of two experiments. Results of each herb type from all time points (total of 30 samples for basil, chive, and cilantro; 36 samples for oregano, parsley, and rosemary) were used for statistical analysis. Means in a row followed by different letters are significantly different by Duncan s multiple range test ( 0.05).

4 2000 HSU ET AL. J. Food Prot., Vol. 69, No. 8 TABLE 3. Effect of storage time on the survival of and on fresh herbs refrigerated at 4 C Population (log CFU/g [fresh wt]) versus storage (day) a Pathogen A 5.68 A 5.16 B 4.86 C 5.16 B 4.60 D 4.91 C 4.43 E 4.80 C 4.14 F 4.43 D a Means reflect the average of two experiments. Results of all test herbs at each time point (total of 36 samples for each storage time except for day 24; 18 samples for day 24) were used for statistical analysis. Means in a row followed by different letters are significantly different by Duncan s multiple range test ( 0.05). O157:H7 and (4.02 and 4.58 log CFU/g, respectively). No statistical differences in populations were observed among chive, oregano, and parsley or chive, oregano, and basil. Survival of on basil, chive, and cilantro was not significantly different. Populations (log CFU per gram) of and recovered from each herb on the first day of storage were 4.96 to 5.65 and 5.42 to 6.06, respectively (Table 1). The initial inoculation level of each bacterium on each sample was approximately 6.30 log CFU/g. The difference in recovery could be the result of bacterial cell death due to drying after inoculation or exposure to antimicrobial components (such as essential oils) released from herb cells during sample processing (3). Antimicrobial activity of plant essential oils has been documented; the essential oil from oregano completely inhibited growth of E. coli O157:H7 and Typhimurium, while the oils from basil, coriander, and rosemary showed a strong inhibitory effect (8). The same researchers demonstrated that essential oil from parsley had the least inhibitory growth effect on both pathogens (8). Burnett and Beuchat (3) compared sample preparation methods for recovering from fresh produce and observed that percent recovery was substantially lower in herbs than in fruits and vegetables. They suggested that the reason for recovering lower numbers was the phenolic compound as well as other compounds that are present in essential oils that are lethal to. The influence of storage time on the survival of test bacteria was also studied (Table 3). A significant (P ) reduction in populations of and was observed within 5 days of storage (0.28 and 0.53 log CFU/g, respectively). After 5 days, decreases in populations varied with pathogen and storage time. Storage time had a significant effect on the survival of E. coli O157:H7, and the population numbers significantly declined with storage time (Table 3). Although populations remained unchanged between 5 and 11 days of storage, significant decreases were detected again at 16 days. The visual quality of herbs was also monitored. The quality of herbs remained good within 11 days of storage. On day 11, the appearance of herbs was unchanged, with only slight discolorations on chive and some small chilling injuries on basil leaves. On day 19, the quality of some herbs started to deteriorate. Parsley leaves had discolorations, and chive, basil, and cilantro showed yellowing on the leaves. Among test herbs, rosemary had the longest shelf life, followed by oregano. Oregano showed slight discolorations on the tips of leaves at day 24, while rosemary remained good in appearance over the entire storage period. At day 24, chive, basil, and cilantro were unusable. Leaf damage (blackened and water-soaked leaves) and visible mold growth were observed on basil. The herbs showing damage developed several signs of deterioration, including severe discoloration and sliminess or fluid leakage. Results suggested that all herbs tested remained of good quality for approximately 2 weeks. Both and E. coli O157: H7 on inoculated fresh herbs survived at least 19 days of storage at 4 C and were very persistent at a refrigerated temperature. Results indicated that cilantro had the highest and rosemary the lowest populations of both test pathogens. Interestingly, cilantro was also identified as one of the herbs that was most often contaminated with or Shigella (9). Although statistically significant, only slight decreases in and populations on each herb were observed, and concentrations remained high over the entire storage period. Results demonstrated that both pathogens were very persistent in all test herbs, even in those stored at 4 C, which is an upper limit of the recommended refrigerated temperature in the FDA Model Food Code These results reinforce the concept that once herbs are contaminated, foodborne pathogens may survive long enough throughout a normal distribution time to pose foodborne illness risks to consumers. Further studies are needed to determine factors affecting the survival rate of these foodborne pathogens on popular culinary herbs that are normally used as edible garnishes. ACKNOWLEDGMENTS This research was supported, in part, by the Office of the Civil Service Commission in Thailand. The authors thank Dr. Tun-Shi Huang of Auburn University for providing strains of and. REFERENCES 1. Abdul-Raouf, U. M., L. R. Beuchat, and M. S. Ammar Survival and growth of Escherichia coli O157:H7 on salad vegetables. Appl. Environ. Microbiol. 59: Brandl, M. T., and R. E. Mandrell Fitness of enterica serovar Thompson in the cilantro phyllosphere. Appl. Environ. Microbiol. 68: Burnett, A. B., and L. R. Beuchat Comparison of sample preparation methods for recovering from raw fruits, vegetables, and herbs. J. Food Prot. 64: Campbell, J. V., J. Mohle-Beoetani, R. Reporter, S. Abbott, J. Farrar, M. Brandl, R. Mandrell, and S. B. Werner An outbreak of

5 J. Food Prot., Vol. 69, No. 8 SURVIVAL OF E. COLI O157:H7 A SALMONELLA ON HERBS 2001 serotype Thompson associated with fresh cilantro. J. Infect. Dis. 183: Centers for Disease Control and Prevention Outbreaks of Shigella sonnei infection associated with eating fresh parsley United States and Canada, July August Morb. Mortal. Wkly. Rep. 48: Croci, L., D. Medici, C. Scalfaro, A. Fiore, and L. Toti The survival of hepatitis A virus in fresh produce. Int. J. Food Microbiol. 73: Czechowicz, S. M., O. Santos, and E. A. Zottola Recovery of thermally stressed Escherichia coli O157:H7 by media supplemented with pyruvate. Int. J. Food Microbiol. 33: Elgayyar, M., F. A. Draughon, D. A. Golden, and J. R. Mount Antimicrobial activity of essential oils from plants against selected pathogenic and saprophytic microorganisms. J. Food Prot. 64: Foley, D., M. Euper, F. Caporaso, and A. Prakash Irradiation and chlorination effectively reduce Escherichia coli O157:H7 inoculated on cilantro (Coriandrum sativum) without negatively affecting quality. J. Food Prot. 67: Hoang, L. M. N., M. Fyfe, C. Ong, J. Harb, S. Champagne, B. Dixon, and J. Isaac-Renton Outbreak of cyclosporiasis in British Columbia associated with imported Thai basil. Epidemiol. Infect. 133: Islam, M., M. P. Doyle, S. C. Phatak, P. Millner, and X. Jiang Persistence of enterohemorrhagic Escherichia coli O157:H7 in soil and on leaf lettuce and parsley grown in fields treated with contaminated manure composts or irrigation water. J. Food Prot. 67: Islam, M., J. Morgan, M. P. Doyle, S. C. Phatak, P. Millner, and X. Jiang Persistence of enterica serovar Typhimurium on lettuce and parsley and in soils on which they were grown in fields treated with contaminated manure composts or irrigation water. Foodborne Pathogen Dis. 1: Kader, A. A Quality and safety factors: definition and evaluation of fresh horticultural crops, p In A. A. Kader (ed.), Postharvest technology of horticultural crops, 3rd ed. University of California Agriculture and Natural Resources Communication Services, Oakland. 14. Karenlampi, R., and M. L. Hanninen Survival of Campylobacter jejuni on various fresh produce. Int. J. Food Microbiol. 97: Lang, M. M., L. J. Harris, and L. R. Beuchat Survival and recovery of Escherichia coli O157:H7,, and Listeria monocytogenes on lettuce and parsley as affected by method of inoculation, time between inoculation and analysis, and treatment with chlorinated water. J. Food Prot. 67: Lopez, A. S., D. R. Dodson, M. J. Arrowood, P. A. Orlandi, Jr., A. J. Silva, J. W. Bier, S. D. Hanauer, R. L. Kuster, S. Oltman, M. S. Baldwin, K. Y. Won, E. M. Nace, M. L. Eberhard, and B. L. Herwaldt Outbreak of cyclosporiasis associated with basil in Missouri in Clin. Infect. Dis. 32: McCabe-Sellers, B. J., and S. E. Beattie. Food safety: emerging trends in foodborne illness surveillance and prevention J. Am. Diet. Assoc. 104: Naimi, T. S., J. H. Wicklund, S. J. Olsen, G. Krause, J. G. Wells, J. M. Bartkus, D. J. Boxrud, M. Sullivan, H. Kassenborg, J. M. Besser, E. D. Mintz, M. T. Osterholm, and C. W. Hedberg Concurrent outbreaks of Shigella sonnei and enterotoxigenic Escherichia coli infections associated with parsley: implications for surveillance and control of foodborne illness. J. Food Prot. 66: Poms, R. E., and S. R. Tatini Survival of Helicobacter pylori in ready-to-eat foods at 4 C. Int. J. Food Microbiol. 63: Simonne, A. H., A. Nille, K. Evans, and M. R. Marshall, Jr Ethnic food safety trends in the United States based on CDC foodborne illness data. Food Prot. Trends 24: Steele, M., and J. Odumeru Irrigation water as source of foodborne pathogens on fruit and vegetables. J. Food Prot. 67: U.S. Department of Agriculture FDA survey of imported fresh produce. U.S. Department of Agriculture, Washington, D.C. Available at: dms/prodsur6.html. Accessed 27 July U.S. Department of Agriculture FDA survey of domestic fresh produce: FY 2000/2001 field assignment. U.S. Department of Agriculture, Washington, D.C. Available at: dms/prodsu10.html. Accessed 27 July Wu, F. M., M. P. Doyle, L. R. Beuchat, J. G. Wells, E. D. Mintz, and B. Swaminathan Fate of Shigella sonnei on parsley and methods of disinfection. J. Food Prot. 63: