Reducing Salmonella on Apples with Wash Practices Commonly Used by Consumers

Transcription

1 741 Journal of Food Protection, Vol. 66, No. 5, 2003, Pages Copyright q, International Association for Food Protection Reducing Salmonella on Apples with Wash Practices Commonly Used by Consumers TRACY L. PARNELL AND LINDA J. HARRIS* Department of Food Science and Technology, University of California, One Shields Avenue, Davis, California , USA MS : Received 21 May 2002/Accepted 18 October 2002 ABSTRACT The ef cacy levels of practices used by consumers to wash smooth-surface fruits and vegetables were compared. Golden Delicious apples were spot inoculated near the blossom end with 50 ml of a cocktail of six serotypes of Salmonella enterica (with a total inoculum level of approximately 10 9 CFU per apple). The inoculum was dried for 1.5 h, and apples were either treated immediately or held for 24 h prior to treatment. Treatments included wetting with approximately 5 ml of water, vinegar (5% acidity), or a 200-ppm chlorine solution, rubbing for 5 or 30 s, rinsing with 200 to 600 ml of 24 or 438C water, and drying with a sterile paper towel. Residual populations of Salmonella were determined by rubbing the treated apple for 30 s in 20 ml of Dey-Engley neutralizing broth and plating on tryptic soy agar and bismuth sul te agar. Rubbing treatments carried out for 5 and 30 s both resulted in a signi cant reduction in Salmonella populations (1 log 10 CFU per apple) relative to populations on samples held for 30 s. A 5-s rub followed by a 200-ml owing-water rinse reduced populations by 3 log 10 CFU per apple. No further decrease in population was obtained by rinsing with 400 or 600 ml of water. Increasing the rinse water temperature to 438C did not signi cantly improve microbial removal. Drying the apple with a sterile paper towel resulted in an additional decrease of approximately 0.4 log 10 CFU per apple. A reduction of 3.2 log 10 CFU was achieved with a combination of wetting with water, rubbing for 5 s, rinsing with 200 ml of water, and drying with a paper towel for apples inoculated just prior to or 24 h before treatment. Reductions obtained for apples treated with 5% vinegar and with a 200-ppm chlorine solution were signi cantly larger (2.1 to 3.2 log 10 CFU per apple, respectively) than those achieved with water. The consumption of fresh fruits and vegetables in the United States increased by 31.5% between 1978 and 1998 (37). This increase has been attributed partly to a heightened awareness that healthier diets include multiple daily servings of fruits and vegetables and to new technologies that have improved product convenience, availability, and shelf life (34). Outbreaks associated with a variety of fresh fruits and vegetables have also increased over the past decade (33). Although vegetables are more often implicated in foodborne outbreaks, fresh fruits such as cantaloupe, tomatoes, and watermelon and fruit juices such as unpasteurized orange juice, apple cider, and apple juice have also been linked to foodborne illness (22). Water, soil, equipment, animals, and humans can all potentially contaminate produce with pathogens (3). Produce can become contaminated during growth or harvest on the farm, during postharvest handling and packing, during transportation, during distribution, during processing, or during nal preparation in a food service establishment or in the home of the consumer. While it is recognized that the elimination of human pathogens from fresh produce is not currently achievable, a reduction of the potential for contamination is desirable. Because contamination can occur at any point from production to nal preparation, the reduction of foodborne illness resulting from the consump- * Author for correspondence. Tel: ; Fax: ; ljharris@ucdavis.edu. tion of fresh produce will require the concerted efforts of all food handlers, from growers to consumers. Some consumer guidance on the handling of fresh produce is available (18, 19). Consumers are reminded to prevent cross-contamination by avoiding the use of the same utensils and surfaces for the preparation of both raw meat and produce. Frequent hand washing and cleaning of surfaces prior to the preparation of fruits and vegetables are emphasized as critical procedures for reducing the risk of foodborne illness. It is also recommended that all produce be rinsed under cool running water just prior to consumption and, when possible, that fruits and vegetables be scrubbed with a clean brush. However, there are few scienti c data available to defend these recommendations or to provide more speci c guidance with regard to the removal of pathogens from the surfaces of fruits and vegetables with techniques commonly available to consumers. The in ltration or uptake of pathogens can occur in dip or soak solutions, particularly when the temperature differential between the produce and the liquid is negative, i.e., when the produce being treated is warmer than the immersion solution (1, 42). This outcome has been demonstrated for apples (9, 11), tomatoes (42), and lettuce (36). Burnett et al. (11) used confocal scanning laser microscopy to evaluate Red Delicious apples that had been immersed in a solution of Escherichia coli O157:H7. The in ltration of cells through the oral tube and their attachment to internal seeds and tissues occurred in spite of a temperature differential between the apple and the inoculum. For these rea-

2 742 PARNELL AND HARRIS J. Food Prot., Vol. 66, No. 5 sons, the soaking of fruits or vegetables during cleaning should be discouraged. In a recent survey, 53% of consumers were found to be willing to try an antibacterial treatment for washing fruits and vegetables at home (32). Most consumers (74%) were not willing to soak apples in a heat treatment with the antimicrobial substance for 15 min. Although consumers are willing to wash produce with an antimicrobial agent, they are interested in processes that require minimal amounts of time and effort. The objective of the present study was to compare the ef cacy levels of apple-washing methods under conditions that would typically exist in consumers homes. Apples were used as a model system for smooth-surface fruits and vegetables because they are consumed in large quantities and are typically eaten raw. Although no outbreaks associated with the consumption of fresh apples have been reported, contaminated apples are thought to be the source of outbreaks of salmonellosis or E. coli O157:H7 gastroenteritis (2, 12 14) associated with unpasteurized apple juice or cider. Salmonella was chosen as a model organism for the current study because its survival on the surfaces of tomatoes was found to be signi cantly better than that of E. coli O157:H7 (6). MATERIALS AND METHODS Fruit preparation. Ripe, waxed Golden Delicious apples were purchased at local grocery stores or produce markets and were not washed or sanitized prior to being used in the experiments. Apples with visible bruises, cracks, or other defects were excluded from the experiments. Apples were stored at,188c for up to 1 week and were brought to room temperature ( C) prior to being used in the experiments. Bacterium cultures. The Salmonella enterica serotypes used in this study were provided by Dr. Larry Beuchat, University of Georgia. Strains had been adapted for growth in the presence of 50 mg of nalidixic acid per ml and were originally isolated from alfalfa sprouts (Salmonella Agona), human feces associated with an egg outbreak (Salmonella Enteritidis E190-88), orange juice (Salmonella Gaminara F2712), cantaloupe (Salmonella Michigan), tomatoes (Salmonella Montevideo G4639), and bovine feces (Salmonella Typhimurium Group B ST). Inoculum preparation. Prior to each experiment, frozen 15% (wt/vol) glycerol stock cultures were streaked onto tryptic soy agar (TSA; Difco Laboratories, Detroit, Mich.) containing 0.1% pyruvic acid (Fisher, Fair Lawn, N.J.) and 50 mg of nalidixic acid (Sigma, St. Louis, Mo.) per ml (TSAPN) and incubated for 24 h at 378C. The addition of pyruvic acid to media has been shown to aid in the recovery of injured cells (26, 30). A single isolated colony was used to inoculate a 10-ml TSA slant containing 50 mg of nalidixic acid per ml without pyruvic acid (TSAN) (20). The entire slant surface was streaked to produce a bacterial lawn and was then incubated for 24 h at 378C. The strains were transferred onto TSAN slants for an additional two consecutive 24-h periods. Cells were collected by adding 5% horse serum albumen (Difco) to each individual culture tube. Horse serum albumen is an organic carrier representing possible sources of contamination, such as soil and feces, that may naturally carry and protect pathogens (5). The surface of the slant was gently scraped with a sterile pipette to remove cells. To prepare a cocktail inoculum, equal amounts of all strains were combined, resulting in a nal concentration of 10 9 CFU/ml. Inoculum levels were con rmed through serial dilutions in Dey-Engley (DE) neutralizing broth (Difco) and the plating of the inoculum onto selective media, TSAPN and bismuth sul te agar (BSA; Difco) with 50 mg of nalidixic acid per ml (BSAN). Inoculation procedure. Apples used in washing experiments were inoculated as described by Beuchat et al. (6). Brie y, apples were positioned with the stem end down and inoculated with 50 ml of cocktail with the application of 10 to 15 drops to the apple skin around the blossom area, with care being taken not to place inoculum directly in the blossom area. Apples were allowed to dry for 1.5 h 6 15 min at C in a biohood with the fan on. The inoculation of apples was staggered to allow time to process all samples within 2 h; however, the inoculum was used within 1 h of preparation. Populations surviving on the surface of the apple after drying were determined for samples receiving no treatment. For each treatment, four to eight apples were evaluated, and each experiment was replicated two to four times. A spot method, rather than immersion in culture broth, was chosen for the inoculation of apples because this method provides a known initial concentration of bacteria, avoids the concentration of bacteria around the blossom end and into the apple core, and allows more effective evaluation of surface decontamination (5, 9). Washing procedures. Apples were processed either immediately after drying for 1.5 h or after storage at ambient temperature for 24 h after drying. Apples were placed in a plastic polyethylene bag (30.5 by 30.5 cm [12 by 12 in.]; Bitran, Com-Pac Int., Carbondale, Ill.), wet with approximately 5 ml ( ml) of sterile distilled water (control), vinegar (5% acidity; H. J. Heinz Co., Pittsburgh, Pa.), or a 200-ppm chlorine solution (sodium hypochlorite solution; Aldrich Chemical Co. Inc., Milwaukee, Wis.). The entire surface of the apple was rubbed by hand for 5 or 30 s or held for 30 s (control). A single person was responsible for rubbing the apples in all experiments in order to reduce variability in the pressure and intensity of the rub. Apples were subjected to a 200-ml distilled-water rinse in one of two ways. For one rinse procedure, the inoculated apple was placed with the blossom (inoculated) end up in a strainer, and 200 ml of sterile 24 or 438C distilled water was poured over the apple (10 cm above the apple, stream diameter ca. 2.5 cm). In some cases, two or three sequential 200-ml volumes of water were poured over the apple. Alternatively, after rubbing, apples were transferred to a new bag containing 200 ml of water, and the bag was agitated for 10 s. Sterile paper towels autoclaved at 1218C for 15 min were sometimes used to remove remaining liquid from the intact apple. Sterile paper towels were aseptically placed in a new plastic bag, the wet apple was placed on the paper towels, and the remaining liquid was wiped off. The 200-ppm chlorine solution was prepared daily with 0.1 M potassium phosphate buffer (ph 6.8, 248C). The total chlorine concentration was tested with a chlorine test kit (Hach Chemical, Loveland, Colo.). Recovery of inoculated cells. Bacteria remaining on the apples were enumerated in one of two ways. In select studies, the inoculated area was excised with a ame-sterilized scalpel, removing skin and as little esh as possible. The skin section was transferred to a 180-ml bag (Whirl-Pak, Modesto, Calif.) with 10 ml of DE neutralizing broth and stomached at the normal setting for 120 s (Stomacher 400, Seward, London, UK). In most cases, apples were placed in a plastic bag with 20 ml of DE neutralizing broth and rubbed vigorously by hand for 30 s (21). When

3 J. Food Prot., Vol. 66, No. 5 REDUCING SALMONELLA ON APPLES 743 TABLE 1. Recovery of Salmonella by excision and rubbing methods from apples inoculated and dried for 1.5 h Estimated inoculum level (log 10 CFU/apple) Salmonella recovered (log 10 CFU/apple) by method a Excision b A C D Rubbing B C D a Average Salmonella population for eight samples from each of two experiments (n 5 16) enumerated on bismuth sul te agar supplemented with 50 mg of nalidixic acid per ml. Values with different letters are signi cantly different (P, 0.05). b Marked squares (2.5 cm 2 ) were excised and stomached or rubbed by hand for 30 s in 20 ml of Dey-Engley neutralizing broth. apples were washed with vinegar or chlorine solution, the DE neutralizing broth and rinse water samples were ltered (10 ml) through a 0.45-mm analytical test lter funnel (Nalgene, Nalge Nunc International, Rochester, N.Y.), rinsed twice with 20 ml of 0.1 M phosphate buffer, and plated on BSAN. Samples were serially diluted in DE neutralizing broth and spread plated on TSAPN and BSAN (0.1 ml on duplicate plates) with the Autoplater 4000 Spiral Plater (Spiral Biotech, Inc., Bethesda, Md.). Plates were counted by hand 24 to 48 h after incubation at 378C according to the guidelines outlined in the Compendium of Methods for the Microbiological Examination of Foods (15). BSAN alone was used for the enumeration of bacteria in experiments involving a water treatment. To determine whether cell injury occurred with a 200-ppm chlorine solution or vinegar treatment, samples were enumerated on both TSAPN and BSAN. The reduction in the bacterial population (log 10 CFU per apple) was calculated as the count for the untreated samples (log 10 CFU per apple) minus the count for the treated samples (log 10 CFU per apple). Colonies were identi ed as Salmonella on the basis of their appearance on BSAN media. Individual colonies were occasionally con rmed with API 20E 24-h test strips (biomerieux Vitex, Inc., Hazelwood, Mo.) or the BBL Entertotube II (Becton Dickinson and Co., Cockeysville, Md.). TABLE 2. Reduction of Salmonella on intact apples achieved by rubbing after wetting with 5 ml of sterile distilled water a Treatment No rub, holding for 30 s (control) 5-s rub 30-s rub Salmonella recovered (log 10 CFU/apple) b A A B B a The estimated inoculum level was log 10 CFU per apple. b Average Salmonella population for four samples from each of four experiments (n 5 16), enumerated on bismuth sul te agar supplemented with 50 mg of nalidixic acid per ml. Values with different letters are signi cantly different (P, 0.05). Statistical analysis. Tukey s post hoc grouping and repeatedmeasures analyses were carried out with the use of PROC GLM in the SAS software package (Version 8, Statistical Analysis Systems Institute, Inc., Cary, N.C.). Differences were considered signi cant at P, RESULTS AND DISCUSSION Inoculation and enumeration of Salmonella on apples. Standard methods for the inoculation and recovery of microorganisms from fruits and vegetables are essential for evaluating the ef cacy of wash treatments (5). The inoculum level used in this study was higher than the perceived natural level on apples, but these higher levels were necessary to allow the observation of decreases in populations after treatment (7). Reductions in populations resulting from these same treatments with a lower inoculum level were not determined. Apples were spot inoculated with Salmonella at three levels (8.9, 5.9, and 4.1 log 10 CFU per apple), and Salmonella were enumerated after the drying period by both excision and rubbing methods (Table 1). A slight but signi cant difference was observed between the counts for the excision and rubbing recovery methods for apples inoculated at the highest bacterium level but not at the two lower levels. The initial number of bacterial cells applied to the apples was larger than the numbers of cells recovered after the drying period by ca. 0.5 to 1 log 10 CFU per apple. Similar reductions during drying were observed for other produce items (data not shown). These differences can be attributed, at least in part, to cell death, although further studies are necessary to verify this conclusion. With the rubbing method, CFU per apple (n 5 16), or 37% of the initial count, was recovered, which is similar to the level of recovery observed by Burnett and Beuchat (10), who showed that washing in 0.1% peptone (41 to 28% recovery) or homogenizing (blending) (34 to 21% recovery) recovered signi cantly larger numbers of Salmonella from Granny Smith and Red Delicious apples than did the stomaching of chopped apples (11 to 14% recovery). Rubbing as a means of recovering Salmonella from apples was considered appropriate for a comparison the ef cacy levels of various wash methods. Evaluation of rubbing intact apples. Populations of Salmonella remaining on apples that had been wetted with 5 ml of sterile distilled water and held for 30 s (control) were equivalent to populations on apples that received no treatment (Table 2). A 5- or 30-s rub after wetting resulted in a reduction of 1 log 10 CFU per apple relative to populations on untreated apples. Reductions observed with a 5- or 30-s rub were signi cantly larger than those for untreated or control apples, but reductions achieved with a 30-s rub were not signi cantly larger than those achieved with a 5- s rub. In a consumer survey, 22% of consumers reported rubbing apples with their hands (for an unknown amount of time) as a preparation method (27). Rubbing an apple for 30 s was considered too long to be practical (8). Evaluation of a water rinse on intact apples. Consumers often rinse produce under running water for short periods as part of a washing protocol (27). The volume of water delivered from a faucet during 5 s of moderate ow was determined to be approximately 200 ml. When apples

4 744 PARNELL AND HARRIS J. Food Prot., Vol. 66, No. 5 TABLE 3. Rinse variables for the removal of Salmonella from intact apples after wetting with 5 ml of sterile distilled water and rubbing for 30 s Variable Effect of rinse Without rinse With rinse Rinse method Agitation Pouring Rinse temperature 248C water rinse 438C water rinse Rinse volume 200 ml 400 ml 600 ml Salmonella recovered a Log 10 CFU/apple A B C L M N S T T X Y Y Y Log 10 CFU/200 ml of rinse water O O U U Z Z Z a Average Salmonella populations for eight samples from each of two experiments (n 5 16), enumerated on bismuth sul te agar supplemented with 50 mg of nalidixic acid per ml., not applicable. Values with different letters are signi cantly different (P, 0.05). were sprayed with 5 ml of water, rubbed for 30 s, and rinsed by agitation in a bag containing 200 ml of sterile distilled water, the decrease observed was signi cantly larger (by 2.3 log 10 CFU per apple) than that achieved with rubbing alone (Table 3). A rinse-based or soak method was evaluated as a means to remove Salmonella from apples. Although a soakrinse method was convenient for containment purposes, there was concern that this method did not accurately re ect rinsing under a kitchen tap and that rinsing in a bag could further contaminate the apple surface. For this reason, apples were sprayed with 5 ml of sterile distilled water, rubbed for 30 s, and rinsed with 200 ml of sterile distilled water by either agitating the apple for 10 s in a bag containing the rinse or pouring an equivalent volume over the apple in a strainer. The pour and agitation rinse methods resulted in reductions of 2.6 and 3.0 log 10 CFU per apple, respectively, a small but signi cant difference (Table 2). Because the pour rinse method is a better model for rinsing produce under running water, it was used in further experiments. Although water pressure was not measured, it was assumed to be approximately similar to that of water owing from a moderately running faucet with a diffuser in place. Warm rinse temperatures are sometimes recommended to consumers for washing produce (24, 25, 39). Under commercial conditions, the maintenance of water soak temperatures that are at least 108C warmer than the fruit pulp temperature decreases the potential for in ltration (1, 4, 42). Rinsing greatly reduces in ltration potential, particularly when rinse times are short and there is little opportunity for the temperature of the fruit to decrease. Water rinses of C and C were compared. The warmer temperature, 438C, was chosen because it is the temperature recommended for hand washing (31) and therefore may also be an appropriate temperature at which to wash produce. Both temperatures resulted in Salmonella reductions of 3.5 log 10 CFU per apple when preceded by wetting with 5 ml of sterile water and a 30-s rub (Table 3). In one study, when apples were immersed in 958C water for 15 s or less, signi cant reductions (5 log CFU/g) in surface populations of E. coli were achieved (17). No measurable reductions were observed with 408C water in the same study; however, apples were neither rubbed nor rinsed. While higher temperatures might be appropriate for commercial use, they are not appropriate or practical for consumers. The focus of the present study was on the reduction of Salmonella on the smooth surface area of the apple and not on the stem and blossom ends, which are known to resist cleaning and are dif cult to access by rubbing (9, 11, 27). Consumers should be made aware that these areas may be contaminated even after washing, and cutting through this area may result in the contamination of the edible esh. Cutting through the stem ends of whole tomatoes inoculated with Salmonella Montevideo resulted in the contamination of the center and bottom of the tomato (29), and the transfer of Salmonella from the rind to the edible esh has been demonstrated for cantaloupe (35). Approximately ml (n 5 20) of water was shown to cling to the surface of the apple after rinsing. With populations of log 10 CFU of Salmonella per ml of rinse water (Table 3), an estimated 30,000 CFU may be present in the liquid clinging to the apple. While these levels might not be detectable at the inoculation levels used in these studies, longer or repeated rinses and towel drying were examined as steps to remove additional bacteria. Increasing the rinse water volume from 200 to 400 or 600 ml did not statistically improve the reduction of Salmonella on the surface of the apple (Table 3). All rinse volumes, in combination with wetting with 5 ml of sterile water and a 30-s rub, resulted in a reduction of 3.5 log 10 CFU per apple. An additional small but signi cant reduction of 0.5 log 10 CFU per apple was achieved when the apple was dried with a sterile paper towel after rubbing for 5 s and rinsing with 200 ml of sterile water (Table 4). The removal of the liquid from the surface of the apple in combination with the friction involved in drying with a paper towel may have resulted in the increased reduction. The friction of paper towels has been shown to make them more effective than cloth for the removal of bacteria from hands (16). The

5 J. Food Prot., Vol. 66, No. 5 REDUCING SALMONELLA ON APPLES 745 TABLE 4. Reduction of Salmonella on intact apples after drying with a paper towel a Treatment Not dried with paper towel Dried with paper towel Salmonella recovered b Log 10 CFU/200 ml Log 10 CFU/apple of rinse water A B C D D a Apples were sprayed with 5 ml of water spray, rubbed for 30 s, and rinsed with 200 ml of water. The estimated inoculum level was log 10 CFU per apple. b Average Salmonella population for eight samples from each of two experiments (n 5 16), enumerated on bismuth sul te agar supplemented with 50 mg of nalidixic acid per ml., not applicable. Values with different letters are signi cantly different (P, 0.05). use of disposable paper towels prevents bacterial cross-contamination of clean surfaces, including clean dishes and hands, that might occur with a used nondisposable towel. Evaluation of combined treatments for removal of Salmonella from apples. Consumers who do wash their produce most commonly use water (27). Consumers surveyed used vinegar, a chlorine solution, or dish detergent for the preparation of fresh fruits and vegetables 1 to 4% of the time (27). Sodium hypochlorite (household bleach), readily available in grocery stores, is not labeled as being for use on produce or other foods, even though it is widely used by the produce industry for the maintenance of water quality. Consumers could prepare a ;200-ppm chlorine solution in their homes by adding 4 ml of bleach (6% sodium hypochlorite) to a liter of water. However, the encouragement of the use of household chemicals beyond their label instructions contradicts other consumer messages regarding pesticide use (38). Of the consumers surveyed, 4% used dish detergent on apples (18), although it is not recommended for use in this manner because it has not been approved for use on food and its ef cacy has not been demonstrated. A prototype produce wash made of generally-recognized-as-safe ingredients reduced populations of Salmonella on tomatoes (6, 20) and apples (21) signi - cantly better than did water alone. However, this wash and similar produce-speci c products are not currently approved by the U.S. Environmental Protection Agency for antimicrobial use. Baking soda and salt water are also occasionally used by consumers to clean fruits and vegetables. Sodium bicarbonate solutions (3%) have little to no antimicrobial effect on Salmonella (data not shown), and their use was not further explored. Applying 5 ml of 200-ppm chlorine solution or vinegar (5% acidity), rubbing for 5 s, rinsing with 200 ml of water, and drying with a paper towel resulted in a reduction of approximately 5.2 to 6.2 log 10 CFU per apple (Table 5), which is signi cantly larger than the reduction achieved with water (3.1 log 10 CFU per apple). There was no signi cant difference between the results obtained with TSAPN and BSAN for the water and chlorine treatments. However, with the vinegar treatment there was a slight but statistically signi cant difference between the results obtained with TSAPN and those obtained with BSAN, indicating that cell injury had occurred. Salmonella counts for the vinegar rinse water were low (14 CFU/ml), and counts for the chlorine rinse water were below the detection limit (,0.1 CFU/ml). The reduction of microorganisms in the rinse water is important in reducing the cross-contamination of other produce items, washing surfaces, and the hands of the washer by the rinse water. TABLE 5. Ef cacy of chemical washing treatments for the removal of Salmonella from inoculated intact apples a Salmonella recovered b Log 10 CFU/apple Log 10 CFU/200 ml of rinse water Treatment TSAPN BSAN TSAPN BSAN Immediately after drying Water Vinegar 200-ppm chlorine solution A B DE D A B DE DE Z Z, W, W 24 h after drying Water Vinegar 200-ppm chlorine solution A C, G DE A C F E Y Y, W, W a Apples were treated immediately after a 1.5-h drying period and after 24 h storage at 248C. Apples were sprayed with 5 ml of treatment solution, rubbed for 5 s, rinsed with 200 ml of water, and dried with a paper towel. The estimated inoculum levels were and log 10 CFU per apple for TSAPN and BSAN, respectively. b Average Salmonella population for four samples from each of three experiments (n 5 12). TSAPN, tryptic soy agar supplemented with 0.1% pyruvic acid and 50 mg of nalidixic acid per ml; BSAN, bismuth sul te agar supplemented with 50 mg of nalidixic acid per ml;, not applicable; one or more samples were below the level of detection (,5 CFU/ml for TSAPN,, 0.1 CFU/ml for BSAN). Values with different letters are signi cantly different (P, 0.05).

6 746 PARNELL AND HARRIS J. Food Prot., Vol. 66, No. 5 In the current study, the Salmonella reduction observed for the chlorine treatment was larger than that achieved with water alone by 2.5 log 10 CFU per apple. In a similar study, when tomatoes were spot inoculated with Salmonella, sprayed with 5 ml of 200-ppm chlorine solution, rubbed for 30 s, and rinsed with 200 ml of sterile water, the reduction observed was larger than that achieved with a water treatment by 0.4 to 0.3 log 10 CFU per tomato (6). When whole tomatoes dip inoculated with Salmonella, E. coli O157:H7, or Listeria monocytogenes were soaked in 110 to 320 ppm of chlorine or were sprayed with 200 to 2,000 ppm of free chlorine and then rinsed with water, reductions increased by 1.5 and 1.9 log CFU/cm 2, respectively, relative to those achieved with water (7, 42). Chlorine provided larger reductions than water alone; however, reduction levels varied among studies, perhaps because of differences in inoculation and recovery methods. Salmonella reductions for apples treated with vinegar were larger than those for apples similarly treated with water by 2.2 log 10 CFU per apple. Vinegar has shown similar reductions of bacteria on intact strawberries and broccoli orets (unpublished data) spot inoculated with Salmonella, agitated for 5 min in vinegar (5% acidity), and rinsed with 200 ml of water. A vinegar treatment reduced Salmonella populations on strawberry or broccoli samples by 2.4 and 2.2 log 10 CFU per sample, respectively, relative to those on samples treated similarly with water. Treatments with vinegar have also been shown to result in increased reductions of other types of organisms on produce. Parsley that had been dip inoculated with Yersinia enterocolitica or Shigella sonnei and then dipped into various concentrations of vinegar (1 to 7.6% acidity) exhibited reductions of.6 log CFU/g after 5 to 30 min (23, 40). Strawberries that had been dip inoculated with E. coli O157:H7 and subsequently dipped for 1 min in 2 or 5% acetic acid exhibited reductions of 1.6 log CFU/g (41). Reductions achieved with vinegar treatments were larger than those achieved with water alone; however, sensory or quality changes in produce, particularly fruits, have not been determined. Evaluation of a 24-h inoculum drying time prior to treatment. A longer inoculum drying time was evaluated to determine whether drying time affected the ef cacy of the wash method. After apples were inoculated and dried for 1.5 h, they were processed either immediately after drying or after holding for an additional 24 h at room temperature ( C). Apples were wetted with 5 ml of water, 200-ppm chlorine solution, or vinegar (5% acidity), rubbed for 5 s, rinsed with 200 ml of sterile distilled water, and dried with a sterile paper towel. After a hold time of 24 h, Salmonella populations on the untreated apples were slightly (0.5 log CFU per apple) but not signi cantly lower than those observed at 0 h (Table 5). Reductions of Salmonella on apples after treatment with vinegar and chlorine were similar at 0 and 24 h. Reductions achieved with vinegar and chlorine treatments were significantly larger than those achieved with water treatment at both 0 and 24 h. Cells were not recovered by plating or ltering from most of the chlorine or vinegar rinse water (,0.1 CFU/ml). CONCLUSIONS Consumers wash fruits and vegetables for a variety of reasons, including the removal of soil, waxes, and pesticides, but.6% of consumers do not routinely wash produce at all (27). Surveys indicate that consumers are willing to wash produce if they feel there is a need to do so (27, 32). However, wash methods must be simple and demand little time. The results of the present study show that simple techniques, including wetting, rubbing, rinsing, and drying, are suf cient to signi cantly reduce surface microorganisms on intact apples. Additional studies involving tomatoes (6, 20) and oranges (unpublished data) indicate that these methods should be applicable to other smooth-surface produce. Although they were not speci cally tested, blossom end and stem areas are more dif cult to clean because they are relatively inaccessible to rubbing (9, 11, 28). Educators may consider it appropriate to remind consumers to avoid cutting through or consuming these areas. Vinegar and chlorine treatments signi cantly improved the reduction of bacteria. However, the effect of vinegar treatments on sensory properties has not been determined, and household bleach is not appropriately labeled for use with produce. ACKNOWLEDGMENTS This research was supported in part by an FDA CFSAN grant (grant no. FD-U ) in collaboration with Dr. Christine Bruhn and Amy Li-Cohen (University of California, Davis). REFERENCES 1. Bartz, J. A., and R. K. Showalter In ltration of tomatoes by aqueous bacterial suspensions. Phytopathology 71: Besser, R. E., S. M. Lett, J. T. Weber, M. P. Doyle, T. J. Barrett, J. G. Wells, and P. M. Grif n An outbreak of diarrhea and hemolytic uremic syndrome from Escherichia coli O157:H7 in freshpressed apple cider. J. Am. Med. Assoc. 269: Beuchat, L. R Pathogenic microorganisms associated with fresh produce. J. Food Prot. 59: Beuchat, L. R Surface decontamination of fruits and vegetables eaten raw: a review. World Health Organization document WHO/FSF/FOS/98.2. World Health Organization, Geneva. 5. Beuchat, L. R., J. M. Farber, E. H. Garrett, L. J. Harris, M. E. Parish, T. V. Suslow, and F. F. Busta Standardization of a method to determine the ef cacy of sanitizers in inactivating human pathogenic microorganisms on raw fruits and vegetables. J. Food Prot. 64: Beuchat, L. R., L. J. Harris, T. E. Ward, and T. M. Kajs Development of a proposed standard method for assessing the ef - cacy of fresh produce sanitizers. J. Food Prot. 64: Beuchat, L. R., B. V. Nail, B. B. Adler, and M. R. S. Clavero Ef cacy of spray application of chlorinated water in killing pathogenic bacteria on raw apples, tomatoes, and lettuce. J. Food Prot. 61: Bruhn, C. (University of California, Davis). Personal communication. 9. Buchanan, R. L., S. G. Edelson, R. L. Miller, and G. M. Sapers Contamination of intact apples after immersion in an aqueous environment containing Escherichia coli O157:H7. J. Food Prot. 62: Burnett, A. B., and L. R. Beuchat Comparison of sample preparation methods for recovering Salmonella from raw fruits, vegetables, and herbs. J. Food Prot. 64: Burnett, S. L., J. R. Chen, and L. R. Beuchat Attachment of

7 J. Food Prot., Vol. 66, No. 5 REDUCING SALMONELLA ON APPLES 747 Escherichia coli O157: H7 to the surfaces and internal structures of apples as detected by confocal scanning laser microscopy. Appl. Environ. Microbiol. 66: Centers for Disease Control and Prevention Salmonella Typhimurium outbreak traced to a commercial apple cider New Jersey. Morb. Mortal. Wkly. Rep. 24: Centers for Disease Control and Prevention Outbreak of Escherichia coli O157:H7 infections associated with drinking unpasteurized commercial apple juice. British Columbia, California, Colorado, and Washington, October Morb. Mortal. Wkly. Rep. 45: Centers for Disease Control and Prevention Outbreaks of Escherichia coli O157:H7 infection and cryptosporidiosis associated with drinking unpasteurized apple juice Connecticut and New York, October Morb. Mortal. Wkly. Rep. 46: Downes, F. P., K. Ito, and American Public Health Association Compendium of methods for the microbiological examination of foods, 4th ed. American Public Health Association, Washington, D.C. 16. Doyle, M. P., K. L. Ruoff, M. Pierson, W. Weinberg, B. Soule, and B. S. Michaels Reducing transmission of infectious agents in the home part I: sources of infection. Dairy Food Environ. Sanit. 20: Fleischman, G. J., C. Bator, R. Merker, and S. E. Keller Hot water immersion to eliminate Escherichia coli O157:H7 on the surface of whole apples: thermal effects and ef cacy. J. Food Prot. 64: Food and Drug Administration FDA advises consumers about fresh produce safety. Available at: tpproduc.html. 19. Food Safety Inspection Service and U.S. Department of Agriculture Does washing food promote food safety? Available at: http: Harris, L. J., L. R. Beuchat, T. M. Kajs, T. E. Ward, and C. H. Taylor Ef cacy and reproducibility of a produce wash in killing Salmonella on the surface of tomatoes assessed with a proposed standard methods for produce sanitizers. J. Food Prot. 64: Harris, L. J., C. A. Pettigrew, and C. H. Taylor Assessment of the microbial ef cacy of a prototype GRAS produce wash on apples, abstr. P19. Abstr. 87th Annu. Meet. Int. Assoc. for Food Prot., Atlanta, Ga International Association for Food Protection, Des Moines, Iowa. 22. Institute of Food Technologists Analysis and evaluation of preventive control measures for the control and reduction/elimination of microbial hazards on fresh and fresh-cut produce. Available at: Karapinar, M., and S. A. Gonul Removal of Yersinia enterocolitica from fresh parsley by washing with acetic acid or vinegar. Int. J. Food Microbiol. 16: Kendall, P Tips on safe handling of fresh produce. Available at: Kershaw, M. E. W., and L. C. Medeiros Safe handling of fruits and vegetables. Available at: ;ohioline/hyg-fact/5000/5353.html. 26. Lee, R. M., and P. A. Hartman Optimal pyruvate concentration for the recovery of coliforms from food and water. J. Food Prot. 52: Li-Cohen, A., and C. M. Bruhn Safety of consumer handling of fresh produce from the time of purchase to the plate: a comprehensive consumer survey. J. Food Prot. 65: Liao, C. H., and G. M. Sapers Attachment and growth of Salmonella Chester on apple fruits and in vivo response of attached bacteria to sanitizer treatments. J. Food Prot. 63: Lin, C. M., and C. I. Wei Transfer of Salmonella Montevideo onto the interior surfaces of tomatoes by cutting. J. Food Prot. 60: McDonald, L. C., C. R. Hackney, and B. Ray Enhanced recovery of injured Escherichia coli by compounds that degrade hydrogen peroxide or block its formation. Appl. Environ. Microbiol. 45: McSwane, D., N. Rue, and R. Linton Essentials of food safety and sanitation, 1st ed. Prentice Hall, Upper Saddle River, N.J. 32. McWatters, K. H., M. P. Doyle, S. L. Walker, A. P. Rimal, and K. Venkitanarayanan Consumer acceptance of raw apples treated with an antibacterial solution designed for home use. J. Food Prot. 65: National Advisory Committee on Microbiological Criteria for Foods Microbiological safety evaluations and recommendations on fresh produce. Food Control 10: Pollack, S. L Consumer demand for fruit and vegetables: the U.S. example. Available at: wrs011/wrs011h.pdf. 35. Suslow, T. V., M. Zunega, J. Wu, L. J. Harris, and T. L. Parnell Potential for transference of inoculated and indigenous bacteria from the non-wounded rind of melons to the interior edible esh, abstr. P071. Abstr. 87th Annu. Meet. Int. Assoc. for Food Prot., Atlanta, Ga International Association for Food Protection, Des Moines, Iowa. 36. Takeuchi, K., and J. F. Frank Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. J. Food Prot. 63: U.S. Department of Agriculture Economic Research Service Vegetables and specialties situation outlook report. U.S. Department of Agriculture, Washington, D.C. 38. U.S. Environmental Protection Agency Citizen s guide to pest control and pesticide safety. Available at: OPPTpubs/CitpGuide/citguide.pdf. 39. Williams, E. B Safe handling of fruits and vegetables. Available at: 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: Yu, K., M. C. Newman, D. D. Archbold, and T. R. Hamilton-Kemp Survival of Escherichia coli O157:H7 on strawberry fruit and reduction of the pathogen population by chemical agents. J. Food Prot. 64: Zhuang, R. Y., L. R. Beuchat, and F. J. Angulo Fate of Salmonella Montevideo on and in raw tomatoes as affected by temperature and treatment with chlorine. Appl. Environ. Microbiol. 61: