Survival of Escherichia Coli O157:H7 and Campylobacter Jejuni in Bottled Purified Drinking Water Under Different Storage Conditions

Home , Campylobacter jejuni

Journal of Food Protection, Vol. 000, No. 000, 0000, Pages 000–000 doi:10.4315/0362-028X.JFP-10-368 Copyright G, International Association for Food Protection

Survival of Escherichia coli O157:H7 and Campylobacter jejuni in Bottled Purified Drinking Water under Different Storage Conditions

HAMZAH M. AL-QADIRI,1* XIAONAN LU,2 NIVIN I. AL-ALAMI,3 AND BARBARA A. RASCO2

1Department of Nutrition and Food Technology, Faculty of Agriculture, The University of Jordan, Amman 11942, Jordan; 2School of Food Science, Box 646376, Washington State University, Pullman, Washington 99164-6376, USA; and 3Water and Environment Research and Study Center, The University of Jordan, Amman 11942, Jordan

MS 10-368: Received 1 September 2010/Accepted 15 October 2010

ABSTRACT ;< Survival of Escherichia coli O157:H7 and Campylobacter jejuni that were separately inoculated into bottled purified drinking water was investigated during storage at 22, 4, and 218uC for 5, 7, and 2 days, respectively. Two inoculation levels were used, 1 and 10 CFU/ml (102 and 103 CFU/100 ml). In samples inoculated with 102 CFU/100 ml, C. jejuni was not detectable (.2-log reduction) after storage under the conditions specified above. E. coli O157:H7 was detected on nonselective and selective media at log reductions of 1.08 to 1.25 after storage at 22uC, 1.19 to 1.56 after storage at 4uC, and 1.54 to 1.98 after storage at 218uC. When the higher inoculation level of 103 CFU/100 ml was used, C. jejuni was able to survive at 22 and 4uC, with 2.25- and 2.17-log reductions observed on nonselective media, respectively. At these higher inoculation levels, E. coli O157:H7 was detectable at 22, 4, and 218uC, with log reductions of 0.76, 0.97, and 1.21 achieved on nonselective media, respectively. Additionally, E. coli O157:H7 showed significant differences in culturability (P , 0.05) on the nonselective and selective culture media under the different storage conditions, with storage at 218uC for 2 days being the treatment most inhibiting. The percentage of sublethal injury of E. coli O157:H7 ranged from ,33 to 75%, indicating that microbial examination of bottled water must be done carefully, otherwise false-negative results or underestimation of bacterial numbers could pose a health risk when low levels of pathogens are present.

The most common and widespread health risk associated with as few as 10 cells able to cause human illness and with drinking water is microbial contamination, resulting in potentially, death (3, 27, 43). Monitoring for the presence of waterborne diseases of high morbidity and mortality rates. At this pathogen in drinking water is essential, since water- least 6 billion cases of gastrointestinal illness are estimated to borne outbreaks due to E. coli O157:H7 still occur (19), and occur annually from contaminated water, especially in several outbreaks due to the presence of this pathogen in children, the elderly, and the immunocompromised (2, 3, drinking water have been recently reported (7, 25, 43). 11, 42). Accordingly, an increase in the demand for treated, Additionally, E. coli O157:H7 can survive in bottled purified bottled drinking water is a direct result of public drinking water stored at room temperature (23, 41), and health concerns about the safety of main supply water, and can readily adapt to, and survive under, different environ- due in part to the recent trend of consuming purified water mental conditions, particularly when exposed to sublethal (20, 23, 31). Many believe that bottled water is safer than stresses related to changes in temperature (i.e., freeze and municipal supplies are, and that bottled water does not cold stresses), availability of nutrients, and exposure to contain microorganisms; however, bottled drinking water is chlorine (43). Accordingly, E. coli O157:H7 might not be rarely absolutely free from microbial contaminants, and it can recovered when exposed to such sublethal injuries (22, 30), support the growth of bacterial pathogens (36). More than and for that reason it is important to determine the presence 50% of Americans drink bottled water, and worldwide sales of sublethally injured E. coli O157:H7 in drinking water, of bottled water exceed $5.7 billion (36). As reported by since injured cells could retain the ability to produce Shiga Rosenberg (36), confirmed outbreaks related to consumption toxin (44) and thus pose a risk for the consumption of of bottled drinking water could be linked to water contaminated bottled water. contamination rather than to source water quality. Campylobacter is the most commonly reported bacterial Enterohemorrhagic Escherichia coli O157:H7 is con- cause of gastrointestinal illness worldwide (7, 16, 28). C. sidered one of the most important foodborne pathogens, jejuni is a food and waterborne pathogen responsible for millions of campylobacteriosis cases (enteritis and diarrhea, * Author for correspondence. Tel: z962-6-5355000, Ext. 22422; Fax: z962- followed by rapid recovery) and causes hundreds of fatalities 6-5300806; E-mail: [email protected]; [email protected]. annually. An estimated 250,000 cases of campylobacteriosis

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:17 1 Cust # JFP-10-368R 0 AL-QADIRI ET AL. J. Food Prot., Vol. 000, No. 000 infection occur annually in the United States (37); 183,961 campylobacter enrichment broth (10, 33) consisting of campylo- laboratory–confirmed campylobacteriosis cases recorded in bacter nutrient broth no. 2 (CM0067, Oxoid, Ltd., Basingstoke, the European Union in 2004 (39); however, the etiology of C. UK) and supplemented with campylobacter growth supplement jejuni remains only partly understood (16). Although (SR0232E, Oxoid, Ltd.). C. jejuni broth was then incubated in campylobacteriosis is generally classified as a self-limiting anaerobic jar at 42uC for 24 h under microaerophilic conditions (10% CO ,5% O , and 85% N ) by using an MGC-Pack- disease (14), it can result in life-threatening sequelae, among 2 2 2 MicroAero culture box (Mitsubishi Gas Chemical America, Inc., which is Guillain-Barre´ syndrome (4). New York, NY). One of the main transmission routes for campylobac- After 24 h incubation, 10 ml of broth of each strain was teriosis is water (24); Campylobacter spp. can survive as transferred to a 50-ml sterile centrifuge tube. The tubes were then viable but nonculturable in drinking water. Waterborne centrifuged for 15 min at 5,000 rpm (3,380 | g) (AccuSpin model outbreaks due to C. jejuni have occurred recently in the 400 bench top centrifuge, Fisher Thermo Scientific, Pittsburgh, PA) United States (7) and Wales (35). Bottled water has been to harvest bacterial cells. To eliminate any effect of broth reported as a possible risk factor for C. jejuni infection (15). components and bacterial metabolites, the resultant pellets were Evans et al. (12) found that Campylobacter spp. infection resuspended in 10 ml of sterile 0.85% (wt/vol) saline solution and due to the consumption of bottled drinking water has a risk centrifuged as before. After the second centrifugation, the factor of 12% as compared with eating chicken (31%) and supernatant was decanted, and the pellets were resuspended in 10 ml of the same sterile, bottled water used for the survival study salad vegetables (21%). Accordingly, C. jejuni could and recentrifuged as above. After centrifugation, the resulting accidentally contaminate bottled drinking water, in which washed pellets were then resuspended in sterile 10-ml aliquots of the it could survive and grow during storage, creating a health same bottled water, corresponding to approximately 108 CFU/ml. risk concern. However, most of research performed to date has Inoculation of water samples for survival studies. Using focused on survival of these pathogens in bottled water appropriate serial decimal dilutions, E. coli O157:H7 and C. jejuni during storage at ambient temperature by using high were separately inoculated into the 500-ml water samples to achieve inoculum levels, 103 to 106 CFU/ml (16, 23, 38, 40, 41). two different inoculum concentrations, 1 and 10 CFU/ml, which is 2 3 Therefore, this current research aimed at investigating the equivalent to 10 and 10 CFU/100 ml, respectively. These survival of E. coli O157:H7 and C. jejuni that were inoculum densities were applied because they might best represent separately inoculated into purified bottled drinking water at the realistic contamination levels of bottled water worldwide. Immediately after inoculation, bacterial viable counts were mea- low inoculum levels (1 and 10 CFU/ml) during storage at sured in duplicate by using the membrane filtration technique (5) 22, 4, and 218uC. described previously. Water samples were then stored at 22, 4, and 218uC for 5, 7, and 2 days, respectively. These storage intervals MATERIALS AND METHODS were selected because some consumers keep unsealed bottled water Water samples. Purified drinking water (treated by reverse at room temperature for 2 to 5 days, whereas others prefer to osmosis, microfiltration, and ozone sterilization) bottled in polyeth- refrigerate bottled water and consume it within a week from ylene terephthalate containers was purchased from a local grocery opening. The 218uC treatment was examined to investigate the store. Water was examined within 2 weeks of its production date. As survival of inoculated bacteria in ice. Uninoculated control water labeled, this type of water contained insignificant amounts of sodium samples were also included in this study, stored as above. bicarbonate and calcium chloride (to preserve fresh taste). To examine the microbial quality of purchased bottled water, all water Recovery of bacteria and culture media. To recover samples used for bacterial inoculation were microbially examined in surviving bacteria, water samples were examined in duplicate by duplicate by using a membrane filtration technique (5). Under the membrane filtration technique (5) detailed above. Water aseptic conditions, 100 ml of drinking water sample was filtered samples were gently mixed with a rolling motion to detach through a gridded, sterile, cellulose–nitrate membrane filter (0.45- adhered bacterial cells from bottles surfaces. R2A agar (Difco, BD, mm pore size, 47-mm diameter; Sartorius Stedim–type filters) under Sparks, MD) was used as a nonselective culture medium to recover partial vacuum. The membrane filters were immediately removed and enumerate E. coli O157:H7. R2A agar is considered a low- with sterile forceps and placed onto solid culture media, with a nutrient medium, suitable for culturing stressed and chlorine- rolling motion to avoid entrapment of air. Microbiological tests tolerant bacteria in drinking water (5, 16, 23). R2A agar plates included total: heterotrophic plate count, coliforms and fecal were incubated at 37uC for 24 to 48 h. m Endo agar LES coliforms, Pseudomonas aeruginosa, and C. jejuni by using (Lawrence Experimental Station; Difco, BD) was used for selective nonselective and selective culture media (5). After this and under enumeration of E. coli O157:H7 (5), with agar plates incubated at aseptic conditions, 500-ml water samples were poured into 1-liter 37uC for 24 h. Preston campylobacter nonselective and selective sterile polyethylene terephthalate bottles and kept tightly sealed at culture media were used to recover and enumerate surviving C. 4uC prior to inoculation. jejuni bacteria (10, 15, 29, 33). Preston campylobacter nonselective agar was prepared according to the manufacturer’s instructions, Bacterial strains and inoculum preparation. All bacterial consisting of campylobacter agar base (CM0689, Oxoid, Ltd.), ATCC strains were obtained from Microbiologics, Inc. (St. Cloud, campylobacter growth supplement (SR0232E, Oxoid, Ltd.), and MN). E. coli O157:H7 ATCC 700728 was cultured and activated lysed horse blood (Remel, Lenexa, KS). Preston campylobacter by inoculating a Kwik-Stik swab (Microbiologics, Inc.) into 50 ml selective agar was prepared with the addition of Preston of tryptic soy broth (Bacto, BD, Franklin Lakes, NJ), and then campylobacter selective supplement (SR0117E, Oxoid, Ltd.). incubated at 37uC for a maximum of 24 h to yield a cell count of Nonselective and selective agar plates were incubated at 42uC 8 9 approximately 10 to 10 CFU/ml. C. jejuni ATCC 29428 was for 48 h under microaerophilic conditions (10% CO2,5% O2, and activated by inoculating the Kwik-Stik swab into 50 ml of 85% N2).

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:19 2 Cust # JFP-10-368R J. Food Prot., Vol. 000, No. 000 SURVIVAL OF E. COLI AND C. JEJUNI IN BOTTLED WATER 0

Statistical analysis. The experiment consisted of three observed when microorganisms are injured. After 5 days of independent replicate trials, and the results were expressed as the storage at 22uC, E. coli O157:H7 remained viable in bottled mean viable counts (CFU per 100 ml) ¡ standard deviation by water and was thus capable of causing a health threat. using Excel software (Microsoft, Corp., Redmond, WA). These The survival times and reduction rates of E. coli data were analyzed by one-way analysis of variance and the O157:H7 in water can vary greatly, depending on the type Student’s t test to evaluate whether significant differences in treatments occurred at P , 0.05 by using MATLAB and Simulink of water, storage temperature, salinity, nutrient availability, R2010a programs (MathWorks, Inc., Natick, MA). Discriminant and concentration and type of initial inocula (23, 32, 41). Kerr 3 function analysis was performed to establish a branched dendro- et al. (23) showed that at the inoculum level of 10 CFU/ml, gram structural model for segregation of water samples, according E. coli O157:H7 could survive up to 63 days (0.56 log) in to storage conditions. Results from the first two experiments were noncarbonated, natural mineral water, 42 days (0.64 log) in used for model establishment, and the results from the third sterile, noncarbonated, natural mineral water, and 14 days experiment were used for model validation. (0.64 log) in sterile, distilled deionized water. At the higher inoculum level of 106 CFU/ml, there was no significant RESULTS AND DISCUSSION difference in the survival times between the three types of Although most of reported waterborne outbreaks have water, with reduction in E. coli O157:H7 counts from ,6.5 to been associated with contaminated public water supply (7, ,2.0 log at the end of the 70-day storage period. 13, 16, 17, 23, 35, 43), several reports indicate a potential Warburton et al. (41) indicated that washing of E. coli health risk from contaminated bottled water (36), mainly O157:H7 cells prior to inoculation into spring water resulted due to postprocess contamination and the ability of in a reduction in the viable counts, which slowly decreased pathogenic microorganisms to survive in bottled water. from 106 CFU/ml, to 103 to 104 CFU/ml over 200 days. Some countries including the United States require bottled However, when pooled, unwashed E. coli O157:H7 inocula water be treated to inactivate pathogens by UV irradiation, were used, viable counts increased by 1 to 2 log within the ozonation, reverse osmosis, or microfiltration (36). How- first 2 days, and then E. coli O157:H7 experienced cycles of ever, these methods do not have any residual disinfectant proliferation and die-off, but survived for more than effect and would not prevent nor inhibit microbial 300 days. This proliferation could be due to trace amounts recontamination and regrowth. Several researchers have of nutrients originating from the nutrient broth(s) that had studied the survival of bacteria in bottled drinking water been used to culture the bacterial cells, which then could stored at ambient temperature by using high inoculation have allowed the extended survival of E. coli O157:H7 in densities 103 to 106 CFU/ml (16, 23, 38, 40, 41), which that study. However, in another study, Moreira et al. (32) might not reflect realistic contamination scenarios. Accord- found that E. coli cells could survive for a maximum 20 days ingly, this was the rationale for examining lower inoculum in mineral water contained in polyvinyl chloride and glass levels (1 and 10 CFU/ml), as well as survival of inoculated bottles, both in sterile water and in the presence of bacteria, at these low densities in purified water with autochthonous bacteria. At the same time, E. coli could minimal nutrients, and in ice. survive for more than 20 days in sterile tap water in glass Microbiological examination of bottled drinking water bottles at 22uC when the initial inoculum ranged between before inoculation and for uninoculated controls indicated 103 and 104 CFU/ml. nonmeasurable bacterial contamination. Table 1 shows the For C. jejuni in the present experiment, more than 2-log mean viable counts (expressed in CFU per 100 ml) for the reductions in viable counts were observed in water inoculated water samples at the different storage conditions. inoculated with 103 CFU/100 ml and stored at 22uC for At the inoculation level of 102 CFU/100 ml, the initial 5 days (Table 1). As observed by others (16, 38), the bacterial loads were 1.1 | 102 and 0.9 | 102 CFU/100 ml survival of C. jejuni in inoculated bottled water depends on for E. coli O157:H7 and C. jejuni, respectively. During several factors including mineral content of the water, storage at 22uC, C. jejuni could not survive (100% lethal storage temperature, strain type, inoculum concentration and injury), whereas E. coli O157:H7 was detected with both type, the presence of autochthonous microflora, concentra- nonselective and selective culture media. An almost 1-log tion of dissolved oxygen, and the procedure used to recover reduction was observed in E. coli O157:H7 viable counts injured bacteria. Tatchou-Nyamsi-Konig et al. (38) inves- after 5 days of storage at 22uC; 1.08- and 1.25-log decreases tigated the survival of C. jejuni in bottled natural mineral were found on nonselective and selective agars, respective- water inoculated with 104 CFU/ml and stored at 25uC for ly. At the inoculum concentration of 103 CFU/100 ml, the 9 days. They observed a rapid decline in C. jejuni viable initial bacterial loads were 1.0 | 103 and 0.89 | 103 CFU/ counts, whereas the highest decay rates were calculated 100 ml for E. coli O157:H7 and C. jejuni, respectively. Both between days 3 and 6, and the average decay constant was of these inoculated bacteria were able to survive at 22uC, but 21.6 ¡ 0.29/day. They indicated that the presence of the viable counts of E. coli O157:H7 were higher than C. autochthonous bacteria in natural mineral water did not jejuni counts were when cultured on nonselective and significantly affect the loss of culturability of C. jejuni, selective agars: 0.76- to 0.96-, and 2.25- to 2.47-log whereas small amounts of organic matter could slow the reductions, respectively. Differences in bacterial recovery loss of culturability and create an environment conducive to between the nonselective and selective culture media were the growth of C. jejuni. observed in the case of E. coli O157:H7; enumeration on Theoretically, the reduced osmolality of purified water selective medium could reduce culturability, as is usually has an adverse effect on the survival of C. jejuni in stored

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:19 3 Cust # JFP-10-368R 0 AL-QADIRI ET AL. J. Food Prot., Vol. 000, No. 000

TABLE 1. Mean viable counts of water stored under different conditionsa Inoculation level of 102 CFU/100 mlb Inoculation level of 103 CFU/100 mlc

Water sample Nonselective media Selective media Nonselective media Selective media

Stored at 22uC for 5 days Inoculated with E. coli d e O157:H7 9.17 ¡ 1.17 (1.08) 6.16 ¡ 0.75 (1.25) 172.18 ¡ 8.91 X (0.76) 110.00 ¡ 3.90 X (0.96) Inoculated with C. jejuni NDf ND 5.00 ¡ 0.89 (2.25) 3.00 ¡ 1.41 (2.47) Stored at 4uC for 7 days Inoculated with E. coli O157:H7 7.00 ¡ 1.41 (1.19) 3.00 ¡1.26 (1.56) 106.00 ¡ 4.6 Y (0.97) 35.33 ¡ 3.56 Y (1.45) Inoculated with C. jejuni ND ND 6.00 ¡ 1.41 (2.17) 3.00 ¡ 1.90 (2.47) Stored at 218uC for 2 days Inoculated with E. coli O157:H7 3.17 ¡ 0.75 (1.54) 1.16 ¡ 0.41 (1.98) 61.00 ¡ 6.36 Z (1.21) 15.33 ¡ 2.16 Z (1.81) Inoculated with C. jejuni ND ND ND ND a Values are the mean of three independent replicate experiments ¡ standard deviation. b Average initial counts (control) for E. coli O157:H7 were 1.1 | 102 and for C. jejuni were 0.90 | 102 CFU/100 ml. Samples were enumerated in duplicate by using a membrane filtration technique. c Average initial counts (control) for E. coli O157:H7 were 1.0 | 103 and for C. jejuni were 0.89 | 103 CFU/100 ml. Samples were enumerated in duplicate by using a membrane filtration technique. d Log reduction (expressed in CFU per 100 milliliters). e At the inoculation level (103 CFU/100 ml), within the same column, mean viable counts of E. coli O157:H7 with different letters are significantly different (P , 0.05). f ND, not detected due to lethal injury. bottled water. Guillou et al. (16) studied the survival of C. in bottled water could provide false-negative results in cases jejuni in bottled mineral water that contained different where the level of contamination is very low. As the mineral contents. They concluded that survival was higher inoculum density used in this study was intentionally low in mineral water with a higher mineral content. Reezal et al. (102 CFU/100 ml) and surviving E. coli O157:H7 counts (34) investigated the effect of low-osmolality nutrient media ranged between 1 and 9 CFU/100 ml, it was not statistically (,130 mOsm) on growth and culturability of Campylobac- valid to calculate significant differences at this level of ter spp. at 4, 25, and 42uC, and found that there was an inoculation. However, at the high inoculation level (103 osmotic threshold affecting the survival and growth of CFU/100 ml), E. coli O157:H7 had apparently greater Campylobacter spp. None of the Campylobacter spp. survival, as shown by greater higher recovery and examined grew in media having osmolalities less than cultivation on nonselective agar (0.97-log reduction) as 130; the number of CFU per milliliter decreased rapidly compared with selective media (1.45-log reduction) for regardless of incubation temperature, whereas no cells were inoculated water stored at 4uC for 7 days. E. coli O157:H7 detected after 24 h. However, growth was observed in media viable counts dropped by 1.21 log in nonselective media having osmolalities greater than 175, with significant and 1.81 log in selective media for water samples stored at populations of microbes recovered throughout the experi- 218uC for 2 days. For water inoculated at 103 CFU/100 ml, ment (up to 96 h) at nongrowth temperatures (25 and 4uC); E. coli O157:H7 showed significant differences in cultur- accordingly, higher osmolality enhanced the survival of ability (P , 0.05), according to storage conditions, with Campylobacter spp. These investigators also demonstrated storage at 218uC for 2 days being the treatment most that the cellular morphology was principally coccoid at low inhibiting, as anticipated. These differences in culturability osmolality (because of osmotic stress) rather than the were observed for both the nonselective and selective conventional rod. culture media (Table 1). Additionally, discriminant function As shown in Table 1, after 7 days of storage at 4uC, E. analysis could segregate samples inoculated with E. coli coli O157:H7 was still detectable (,10 CFU/100 ml) in O157:H7 (103 CFU/100 ml) by storage condition. Clear samples inoculated with 102 CFU/100 ml; cells were segregation was achieved for the validated model and the cultivable on both nonselective and selective media. When control, presenting a distinct segregation from all three types samples were stored at 218uC for 2 days, E. coli O157:H7 of storage conditions and providing a quantifiable indication was still able to survive (#3 CFU/100 ml), with a ,2.0-log of the degree of cell injury (Figs. 1 and 2). The interclass reduction observed with selective media. Accordingly, distances ranged from 62.5 to 1,123.9, based on Mahala- using selective media to recover sublethally injured cells nobis distance measurements computed between different

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:19 4 Cust # JFP-10-368R J. Food Prot., Vol. 000, No. 000 SURVIVAL OF E. COLI AND C. JEJUNI IN BOTTLED WATER 0

FIGURE 1. A composite dendrogram de- rived from discriminant function analysis showing differentiations in storage conditions for bottled water inoculated with Escherichia coli O157:H7 at the concentration of 103 CFU/100 ml by using nonselective culture media.

classes. Groups with distance values .3 are considered maximum survival was achieved at 8uC and a minimum significantly different from one another (9), and this at 25uC, regardless of the water source, and they observed a paralleled the significant differences determined by the reduction of viable E. coli O157:H7 by 1 to 2 log by 91 days Student’s t test (P , 0.05), as presented in Table 1. at 8uC; however, viable counts were not detectable (.3-log Obviously, E. coli O157:H7 can pose a health risk as an reduction) within 49 to 84 days at 25uC. According to the accidental contaminant in bottled water and even at very Wang et al. study (40), there was no major alteration in low contamination levels, as it was able to survive during lipopolysaccharide pattern; however, the outer membrane reasonable times of storage and under cold temperatures in protein composition was changed. In a somewhat similar bottled water with essentially no nutrient(s). Wang et al. study, Czajkowska et al. (8) studied the survival of E. coli (40) investigated the survival of a mixture of five nalidixic O157:H7 in river and lake water samples stored at 6uCby acid–resistant E. coli O157:H7 strains (103 CFU/ml) at 8uC using high initial inocula (108 to 109 CFU/ml), and showed in filtered, autoclaved municipal water; reservoir water; and that E. coli O157:H7 was reduced to an undetectable level lake water. They showed that E. coli O157:H7 could survive (,1 CFU/ml) within 32 and 51 days, respectively. for long periods in water, even at cold temperatures, with Survival of C. jejuni at cold temperatures was low, with optimal survival in filtered, autoclaved municipal water and the microbe being undetectable at 4uC in samples inoculated lower survival in lake water. They concluded that a with 102 CFU/100 ml and at 218uC at inoculation levels of

FIGURE 2. A composite dendrogram de- rived from discriminant function analysis showing differentiations in storage conditions for bottled water inoculated with Escherichia coli O157:H7 at the concentration of 103 CFU/100 ml by using selective culture media.

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:20 5 Cust # JFP-10-368R 0 AL-QADIRI ET AL. J. Food Prot., Vol. 000, No. 000

TABLE 2. Percentage of lethal and sublethal injury of Escherichia coli O157:H7 inoculated into bottled water during storage under different conditions Inoculation level of 102 CFU/100 ml Inoculation level of 103 CFU/100 ml

Storage conditions % lethal injury % sublethal injurya % lethal injury % sublethal injurya

22uC for 5 days 91.7 32.8 82.8 36.1 4uC for 7 days 93.6 57.1 89.4 66.7 218uC for 2 days 97.1 63.4 93.9 74.9 a %sublethal injury ~ counts on nonselective media – counts on selective media/counts on nonselective media | 100. either 102 or 103 CFU/100 ml (Table 1). However, C. jejuni ranged between 32.8 and 63.4, and 36.1 and 74.9 for low was able to survive during storage at 4uC in samples and high inoculation densities, respectively. However, the inoculated with 103 CFU/100 ml, with an observed percentage of lethal injury was higher among samples reduction of 2.17 to 2.47 log on nonselective and selective inoculated with 102 CFU/100 ml. Sublethally injured cells agars, respectively. At this inoculum level, no differences in lose their ability to grow on selective media; injured bacteria culturability or survival were observed for water stored at 22 become sensitive to many ingredients present in selective and 4uC. While freezing provides the greatest stress to C. media, which can cause damage to membranes and modify jejuni (21), survival of Campylobacter spp. in aquatic cell permeability (6, 21). Accordingly, sublethally injured environments is usually better at lower temperatures (4 to bacteria are defined as bacteria that are able to grow and 6uC) than it is at higher ones. The temperature threshold at form colonies on nonselective culture media but not on which survival becomes significantly affected is acknowl- selective media (21). As the percentage of sublethal injury edged at 16 to 22uC (16, 38). Tatchou-Nyamsi-Konig et al. among E. coli O157:H7 in bottled water was more than (38) indicated a decline in C. jejuni culturability (0.2 to 0.8 30%, and under all studied storage conditions, it is not log CFU in 9 days) in mineral water stored at 4uC and when recommended to directly use selective media to recover initial inoculum was 104 CFU/ml, with the average decay injured cells in bottled water, because this could signifi- constant at 4uC being 20.11 ¡ 0.045/day. Hazeleger et al. cantly underestimate the contamination levels or lead to (18) reported that many vital physiological activities false-negative results when low levels of pathogens are including catalase activity, ATP synthesis, respiration, and present. protein biosynthesis were observed at temperatures below In conclusion, bottled drinking water can pose a health the minimal growth temperature (30uC) for C. jejuni, threat if accidentally contaminated by bacterial pathogens. indicating that metabolic processes were still active. Although culturability of E. coli O157:H7 declined Aerotaxis and chemotaxis were also observed at all gradually in bottled purified water, it was able to survive temperatures between 4 and 40uC (18). Guillou et al. (16) and was detected on both selective and nonselective culture found that the culturability of C. jejuni inoculated into media during storage at 22, 4, and 218uC. However, C. bottled water with high and low mineral concentrations jejuni was more sensitive to cold stress under these 6 (initial inoculum .10 CFU/ml) progressively reduced conditions as compared with E. coli O157:H7. C. jejuni when samples were stored at 4uC within 48 days (more did not survive in bottled purified water when inoculated at than 6-log CFU/ml reduction in viable counts). They low densities (102 CFU/100 ml), and was undetectable concluded that the loss in culturability was observed at all when samples stored at 218uC. conditions tested and was dependent on strain, preculture conditions, and water composition, although a higher ACKNOWLEDGMENTS mineral content appeared to improve survivability. This research was supported in part by Washington State University; Bacterial growth in bottled water tends to be higher in the University of Jordan, Amman, Jordan; and the U.S. Department of noncarbonated waters and in water bottled in plastic Agriculture. containers as compared with glass bottles (36). Plastic is more permeable to oxygen and extraneous vapors; addition- REFERENCES ally, plastic may release nutrients that could increase bacterial 1. Al-Qadiri, H. M., N. I. Al-Alami, M. A. Al-Holy, and B. A. Rasco. growth. Warburton et al. (41) showed that the adherence of 2008. Using Fourier transform infrared (FT-IR) absorbance spectros- bacteria to bottle surfaces might aid in the survival process. copy and multivariate analysis to study the effect of chlorine-induced = Bacteria are often found attached to container walls, where it bacterial injury in water. J. Agric. Food Chem. 56:8992–8997. 2. Al-Qadiri, H. M., M. A. Al-Holy, M. Lin, N. I. Alami, A. G. is believed that the nutrient density is higher as compared Cavinato, and B. A. Rasco. 2006. Rapid detection and identification with the bulk liquid phase. In the present experiment, swab of Pseudomonas aeruginosa and Escherichia coli as pure and mixed tests were performed to examine the internal surface of water cultures in bottled drinking water using Fourier transform infrared containers by using nonselective culture media; however, spectroscopy and multivariate analysis. J. Agric. Food Chem. 54: bacterial cells were not detected. 5749–5754. 3. Al-Qadiri, H. M., M. Lin, A. G. Cavinato, and B. A. Rasco. 2006. Table 2 presents the percentages of lethal and sublethal Fourier transform infrared spectroscopy, detection and identification injury of E. coli O157:H7 in bottled water under different of Escherichia coli O157:H7 and Alicyclobacillus strains in apple storage conditions. The percentages of sublethal injury juice. Int. J. Food Microbiol. 111:73–80.

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:35 6 Cust # JFP-10-368R J. Food Prot., Vol. 000, No. 000 SURVIVAL OF E. COLI AND C. JEJUNI IN BOTTLED WATER 0

4. Altekruse, S., N. Stern, P. Fields, and D. Swerdlow. 1999. 24. Koenraad, P. M. F. J., F. M. Rombouts, S. H. W. Notermans. 1997. Campylobacter jejuni—an emerging foodborne pathogen. Emerg. Epidemiological aspects of thermophilic Campylobacter in water- Infect. Dis. 5:28–35. related environments: a review. Water Environ. Res. 69:52–63. 5. American Public Health Association, American Water Works 25. Lee, S. H., D. A. Levy, G. F. Craun, M. J. Beach, and R. L. Calderon. Association, and Water Environment Federation. 1998. Standard 2002. Surveillance for waterborne-disease outbreaks—United States, methods for the examination of water and wastewater, 20th ed. 1999–2000. MMWR CDC Surveill. Summ. 51:1–47. American Public Health Association, American Water Works 26. Legnani, P., E. Leono, S. Rapuano, D. Turin, and C. Valenti. 1999. Association, and Water Environment Federation, Washington, DC. Survival and growth of Pseudomonas aeruginosa in natural mineral 6. Besse, N. G. 2002. Influence of various environmental parameters water: a 5-year study. Int. J. Food Microbiol. 53:153–158. and of detection procedures on the recovery of stressed L. 27. Li, J., and C. J. Hovde. 2007. Expression profiles of bovine genes in monocytogenes: a review. Food Microbiol. 19:221–234. the rectoanal junction mucosa during colonization with Escherichia 7. Bopp, D. J., B. D. Sauders, A. L. Waring, J. Ackelsberg, N. Dumas, coli O157:H7. Appl. Environ. Microbiol. 73:2380–2385. E. Braun-Howland, D. Dziewulski, B. J. Wallace, M. Kelly, T. Halse, 28. Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. K. A. Musser, P. F. Smith, D. L. Morse, and R. J. Limberger. 2003. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness Detection, isolation, and molecular subtyping of Escherichia coli and death in the United States. Emerg. Infect. Dis. 5:607–625. O157:H7 and Campylobacter jejuni associated with a large 29. Merino, F. J., A. Agulla, P. A. Villasante, A. Diaz, J. V. Saz, and A. waterborne outbreak. J. Clin. Microbiol 41:174–180. C. Velasco. 1986. Comparative efficacy of seven selective media for 8. Czajkowska, D., A. Witkowska-Gwiazdowska, I. Sikorska, H. isolating Campylobacter jejuni. J. Clin. Microbiol. 24:451–452. Boszczyk-Maleszak, and M. Horoch. 2005. Survival of Escherichia 30. Mizunoe, Y., S. N. Wai, A. Takade, and S. Yoshida. 1999. coli serotype O157:H7 in water and in bottom-shore sediments. Pol. Restoration of culturability of starvation-stressed and low-tempera- J. Environ. Stud. 14:423–430. ture-stressed Escherichia coli O157 cells by using H2O2-degrading 9. De Maesschalck, R., A. Candofi, D. L. Masart, and S. Heuerding. compounds. Arch. Microbiol. 172:63–67. 1999. Decision criteria for soft independent modeling of class analogy 31. Moor, J. E., M. McCalmont, J. Xu, B. C. Millar, and N. Heaney. applied to near infrared data. Chemometr. Intell. Lab. Syst. 47:65–77. 2002. Asaia sp., an unusual spoilage organism of fruit-flavored 10. Donnison, A. 2003. Isolation of thermotolerant Campylobacter— bottled water. Appl. Environ. Microbiol. 68:4130–4131. review and methods for New Zealand laboratories. Prepared for the 32. Moreira, L., P. Agostlnho, P. V. Morals, and M. S. da Costa. 1994. Ministry of Health. Available at: http://www.moh.govt.nz/moh.nsf/ Survival of allochthonous bacteria in still mineral water bottled in > pagesmh/2588. polyvinyl chloride (PVC) and glass. J. Appl. Bacteriol. 77:334–339. 11. Emde, K. M. E., and G. R. Finch. 1991. Detection and occurrence of 33. Peyrat, M. B., C. Soumet, P. Maris, and P. Sanders. 2008. Recovery waterborne bacterial and viral pathogens. Res. J. WPCF 63:730–734. of Campylobacter jejuni from surfaces of poultry slaughterhouses 12. Evans, M. R., C. D. Ribeiro, and R. L. Salmon. 2003. Hazards of after cleaning and disinfection procedures: analysis of a potential healthy living: bottled water and salad vegetables as risk factors for source of carcass contamination. Int. J. Food Microbiol. 124:188– Campylobacter infection. Emerg. Infect. Dis. 9:1219–1225. 194. 13. Ferroni, A., L. Nguyen, B. Pron, G. Quesne, M.-C. Brusset and P. 34. Reezal, A., B. Mcneil, and J. G. Anderson. 1998. Effect of low- Berche. 1998. Outbreak of nosocomial urinary tract infections due to osmolality nutrient media on growth and culturability of Campylo- Pseudomonas aeruginosa in a paediatric surgical unit associated with bacter species. Appl. Environ. Microbiol. 64:4643–4649. tap-water contamination. J. Hosp. Infect. 39:301–307. 35. Richardson, G., D. R. Thomas, R. M. M. Smith, L. Nehaul, C. D. 14. Figueroa, G., M. Troncoso, C. Lopez, P. Rivas, and M. Toro. 2009. Ribeiro, A. G. Brown, and R. L. Salmon. 2007. A community Occurrence and enumeration of Campylobacter spp. during the outbreak of Campylobacter jejuni infection from a chlorinated public processing of Chilean broilers. BMC Microbiol. 94:1–6. water supply. Epidemiol. Infect. 135:1151–1158. 15. Gillespie, I. A., S. J. O’Brien, J. A. Frost, G. K. Adak, P. Horby, A. V. Swan, M. J. Painter, and K. R. Neal. 2002. A case–case comparison of 36. Rosenberg, F. A. 2003. The microbiology of bottled water. Clin. Campylobacter coli and Campylobacter jejuni infection: a tool for Microbiol. 25:41–44. generating hypotheses. Emerg. Infect. Dis. 8:937–942. 37. Stafford, R. J., P. J. Schluter, M. D. Kirk, and A. Wilson. 2009. 16. Guillou, S., I. Leguerinel, N. Garrec, M. A. Renard, J. M. Cappelier, Population-attributable risk estimates for risk factors associated with and M. Federighi. 2008. Survival of Campylobacter jejuni in mineral Campylobacter infection, Australia. Emerg. Infect. Dis. 15:850–852. bottled water according to difference in mineral content: application (Letter to the editor.) of the Weibull model. Water Res. 42:2213–2219. 38. Tatchou-Nyamsi-Konig, J.-A., A. Moreau, M. Federighi, and J.-C. 17. Hanninen, M.-L., H. Haajanen, T. Pummi, K. Wermundsen, M.-L. Block. 2007. Behaviour of Campylobacter jejuni in experimentally Katila, H. Sarkkinen, I. Miettinen, and H. Rautelin. 2003. Detection contaminated bottled natural mineral water. J. Appl. Microbiol. 103: and typing of Campylobacter jejuni and Campylobacter coli and 280–288. analysis of indicator organisms in three waterborne outbreaks in 39. Vandeplas, S., C. Marcq, R. D. Dauphin, Y. Beckers, P. Thonart, and Finland. Appl. Environ. Microbiol. 69:1391–1396. A. Thewis. 2008. Contamination of poultry flocks by the human 18. Hazeleger, W. C., J. A. Wouters, F. M. Rombouts, and T. Abee. 1998. pathogen Campylobacter spp. and strategies to reduce its prevalence Physiological activity of Campylobacter jejuni far below the minimal at the farm level. Biotechnol. Agron. Soc. Environ. 12:317–334. growth temperature. Appl. Environ. Microbiol. 64:3917–3922. 40. Wang, G., and M. P. Doyle. 1998. Survival of enterohemorrhagic 19. Hrudey, S. E., and E. J. Hrudey. 2004. Safe drinking water: lessons Escherichia coli O157:H7 in water. J. Food. Prot. 61:662–667. from recent outbreaks in affluent nations. IWA Publishing, London. 41. Warburton, D. W., J. W. Austin, B. H. Harrison, and G. Sanders. 20. Hunter, P. R. 1993. The microbiology of bottled natural mineral 1998. Survival and recovery of Escherichia coli O157:H7 in waters: a review. J. Appl. Bacteriol. 74:345–352. inoculated bottled water. J. Food Prot. 61:948–952. 21. Jasson, V., M. Uyttendaele, A. Rajkovic, and J. Debevere. 2007. 42. World Health Organization. 1996. Guidelines for drinking water Establishment of procedures provoking sub-lethal injury of Listeria quality, 2nd ed., vol. 2. World Health Organization, Geneva. monocytogenes, Campylobacter jejuni and Escherichia coli O157 to 43. Yanming, L., A. Gilchrist, J. Zhang, and X. Li. 2008. Detection of serve method performance testing. Int. J. Food Microbiol. 118:241–249. viable but nonculturable Escherichia coli O157:H7 bacteria in 22. Juhna, T., D. Birzniece, and J. Rubulis. 2007. Effect of phosphorus drinking water and river water. Appl. Environ. Microbiol. 74:1502– on survival of Escherichia coli in drinking water biofilms. Appl. 1507. Environ. Microbiol. 73:3755–3758. 44. Yaron, S., and K. R. Matthews. 2002. A reverse transcriptase- 23. Kerr, M., M. Fitzgerald, J. J. Sheridan, D. A. McDowell, and I. S. polymerase chain reaction assay for detection of viable Escherichia Blair. 1999. Survival of Escherichia coli O157:H7 in bottled natural coli O157:H7: investigation of specific target genes. J. Appl. mineral water. J. Appl. Microbiol. 87:833–841. Microbiol. 92:633–640.

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:37 7 Cust # JFP-10-368R 0 AL-QADIRI ET AL. J. Food Prot., Vol. 000, No. 000

Authors Queries Journal: Journal of Food Protection Paper: food-74-02-22 Title: Survival of Escherichia coli O157:H7 and Campylobacter jejuni in Bottled Purified Drinking Water under Different Storage Conditions

Dear Author During the preparation of your manuscript for publication, the questions listed below have arisen. Please attend to these matters and return this form with your proof. Many thanks for your assistance

Query Query Remarks Reference

1 Author: This article has been edited for grammar, style, and usage. Please compare it with your original document and make corrections on these pages. Please limit your corrections to substantive changes that affect meaning. If no change is required in response to a question, please write ‘‘OK as set’’ in the margin. Copy editor

2 Author: Key words: Bottled water, drinking water, E. coli O157:H7, C. jejuni, bacterial survival. Key words are used for indexing; please approve or edit as needed. Copy editor

3 Author: References have been renumbered per style for alphabet- ic order; please check references and in-text citations carefully. Copy editor

4 Author: Please cite date Web page was last accessed, in the following format: day month year. Copy editor

Journal of Food Protection food-74-02-22.3d 16/11/10 15:20:38 8 Cust # JFP-10-368R