1860 Journal o f Food Protection, Vol. 77, No. 11, 2014, Pages 1860-1867 doi: 10.4315/0362-028X.JFP-14-220

Behavior of Shiga Toxigenic Escherichia coli Relevant to Lettuce Washing Processes and Consideration of Factors for Evaluating Washing Process Surrogates KAIPING D EN G ,'| XUE WANG,2 LI-HAN YEN ,2 HONGLIU DING , 1 AND MARY LOU TORTORELLO1* 'U.S. Food and Drug Administration, Division o f Food Processing Science and Technology, and institute fo r Food Safety and Health, Illinois Institute o f Technology, 6502 South Archer Road, Bedford Park, Illinois 60501, USA MS 14-220: Received 9 May 2014/Accepted 2 July 2014

ABSTRACT Postharvest processes for fresh produce commonly include washing in water containing antimicrobial chemicals, such as chlorine; however, if the antimicrobials are not present in sufficient levels, washing can promote the spread of contamination that might be present. To understand cross-contamination risk during washing, we tested a collection of Shiga toxigenic Escherichia coli (STEC), including 0157:H7 and other non-0157 strains, for certain traits during washing of fresh-cut lettuce, i.e., sensitivity to sublethal chlorine levels and ability to cross-contaminate (detach from and attach to) lettuce in the presence of sublethal chlorine levels. Nonpathogenic E. coli Nissle 1917 (EcN) and Pediococcus pentosaceus lactic acid bacterial species (LAB) were included as potential washing process validation surrogates. As measured by extension of the lag phase of growth in media containing 0.15 ppm of chlorine, chlorine sensitivity varied among the STECs. Cross-contamination was assessed by evaluating transfer of bacteria from inoculated to uninoculated leaves during washing. Without chlorine, similar transfer to wash water and uninoculated leaves was shown. In 1 ppm of chlorine, cross-contamination was not detected with most strains, except for the substantial transfer by a STEC 0111 strain and EcN in some replicates. Strain 0111 and EcN showed less inactivation in 0.25 ppm of chlorine water compared with 0157 (P < 0.05). LAB showed similar transfer and similar chlorine inactivation to 0157. Considering together the sublethal chlorine sensitivity and detachment/attachment traits, neither EcN nor LAB displayed optimal characteristics as washing process surrogates for the STEC strains, although further evaluation is needed. This work demonstrated a range of behaviors of STEC strains during lettuce washing and may be helpful in hazard characterization, identifying factors to consider for evaluating washing process efficacy, and identifying phenotypic traits to select surrogates to validate washing processes.

Contamination of fresh produce with microbial patho­ gens can occur in various ways during preharvest stages, for example, by application of manures and other soil amendments, exposure to contaminated water, and through contact with animals, including humans ( 17, 39 ). The contamination can persist through postharvest processes and be passed along to consumers, as evidenced by the multiple instances of foodbome illness outbreaks associated with fresh produce ( 12, 52). Antimicrobial chemicals, such as chlorine, are com­ monly used in fresh produce washing processes to reduce the number of microorganisms in production and help to maintain product quality (2). It has long been recognized that antimicrobial washes are minimally effective for decontaminating the product ( 1 , 4 , 40), but when applied appropriately during washing, the antimicrobials can be very effective at preventing cross-contamination within * Corresponding author. Tel: 708-924-0645; Fax: 708-924-0690. E-mail: [email protected]. t Present address: Institute for Food Safety and Health, Illinois Institute of Technology, 6502 South Archer Road, Bedford Park, IL 60501, USA.

the batch ( 17, 20 , 39). Chlorine is a commonly used antimicrobial chemical for produce washing, but it becomes less effective in the presence of organic matter, which is contributed by soil adhering to the product and by the product itself (44). Fresh-cut products, such as cut lettuce, can exude organic leaf constituents into the wash water and severely challenge chlorine’s effectiveness ( 37). Increasing levels of organic matter accumulate, as the production proceeds throughout the day and as the processing water is recycled for reuse ( 19, 44 , 48). The antimicrobial chemical concentration in the wash water can fluctuate and decline to sublethal levels (37 , 44 , 49). If it decreases to levels that are not sufficient to inactivate the pathogens, the potential spread of contamination to equipment and throughout the production batch ( 7, 8) becomes a concern, leading to increased risk to consumers. Antimicrobial levels need to be monitored and readjusted throughout processing to ensure that effective levels are maintained (40 , 41, 44). Despite the existence of automated monitoring and feedback loop adjustments to effective antimicrobial chemical levels, populations of microorganisms, including coliforms, can fluctuate in the wash water during a production day (3 ). The

J. Food Prot., Vol. 77, No. 11

U.S. Food and Drug Administration (FDA) proposes that hazards that are reasonably likely to occur in a food should be controlled in validated processes (46 )\ thus, antimicrobial washes should demonstrate the ability to control cross­ contamination. The ability of pathogens to survive in wash water containing sublethal chlorine levels and to transfer from contaminated leaves to clean leaves are key elements that can contribute to the spread of contamination during fresh produce washing. Compared with the extensively studied response of 0157:H7 to antimicrobials ( 22 , 25, 32), including transfer in antimicrobial-containing washes (56), fewer studies have been published regarding antimicrobial treatment of non-0157 Shiga toxigenic Escherichia coli (STEC) strains, although they have caused illnesses associated with fresh produce consumption (5, 9 , 11, 45). The studies have focused on biofilm removal (50), sanitization of food contact surfaces ( 28), and beef surface treatments ( 13, 18, 24). The survival of non-0157 STEC in solutions of food-grade antimicrobials has been studied to a lesser extent. For example, 026 and 0157 showed similar MICs of 200 ppm of peroxyacetic acid, 0.1% lactic acid, and 0.5% sodium hypochlorite ( 43), and 026, 045, 0103, and 0145 showed greater inactivation kinetics in chlorinated water than 0157:H7 (41). Our objective was to examine the behavior of different STEC strains during processes relevant to fresh-cut lettuce washing, including the ability of strains to survive and cross­ contaminate the product. Transfer from contaminated lettuce leaves to wash water ( 56) and attachment to leaf surfaces (26) have been studied as relevant behaviors in the survival of Escherichia coli 0157:H7 in fresh produce processes. Understanding the behavior of the pathogens is useful for hazard characterization, but it is also necessary to eventually identify surrogate strains for use by the produce industry to validate the effectiveness of antimicrobial washing processes. A surrogate strain should be nonpathogenic, but mimic pathogen behavior, and should be selected by considering the traits relevant to that behavior for the process being evaluated ( 10, 35, 42). Because the essential function of the antimicro­ bial in the wash water is to prevent cross-contamination ( 19, 30, 32), the traits relevant to cross-contamination should be compared in strains of the pathogen and potential siurogate. In lettuce washing processes such traits would include (i) relative sensitivity to antimicrobial chemicals, e.g., chlorine, including concentrations that have declined to sublethal levels due to process effects; (ii) the ability to transfer from contaminated lettuce to wash water, i.e., detachment from lettuce; and (iii) the ability to transfer from wash water to uncontaminated leaves, i.e., attachment to lettuce; and (iv) inactivation in chlorinated solution. Our objectives were to compare these four traits among various STEC strains, as well as potential surrogate strains. Characterizing the range of behaviors among the STEC strains would help to ensure effective preventive controls and, ultimately, to select appropriate surrogate strains for validating antimicrobial wash processes. MATERIALS AND METHODS Bacterial strains. The 36 STEC strains selected for this study included six 0157:H7 isolates from different outbreaks, a set of

STEC BEHAVIOR DURING LETTUCE WASHING

1861

non-0157 “ Big Six” serotypes from the STEC Center in Michigan State University, and other non-0157 serotypes from the FDA culture collection (Table 1). The lettuce outbreakassociated 0157:H7 strain H I827 (24) was provided by Peter Gemer-Smidt (Centers for Disease Control and Prevention, Atlanta, GA). E. coli strain Nissle 1917 (EcN; Ardeypharm, Herdecke, Germany), which is commercially marketed in Europe as a probiotic to treat intestinal disorders (33), was included as a potential surrogate strain. All of the E. coli strains were grown at 37°C in brain heart infusion medium (BHI; BD, Franklin Lakes, NJ). Nalidixic acid (nal; 50 pg/ml; Sigma-Aldrich Co., St. Louis, MO) or chloramphenicol (cm; 30 pg/ml; Sigma-Aldrich Co.) was supplemented when necessary. The lactic acid bacterial species Pediococcus pentosaceus lactic acid bacterial species (LAB) was included as a potential surrogate and was grown anaerobically in Lactobacilli de Man Rogosa Sharpe (MRS) medium (BD) by using a GasPak EZ Anaerobic Pouch System (BD). All strains were stored in a MicroBank Advanced Bacterial Storage System (PRO­ LAB Diagnostics, Round Rock, TX) at -80°C . The strains were propagated from frozen stock by inoculating a streak plate, and after a 24-h incubation of the plate at 37°C, a single colony was inoculated into 2 ml of BHI or MRS broth. After 18- to 24-h incubation, the culture was used for the experiments. An antibiotic marker was introduced into the strains involved in the transfer assays for differentiating the strains from the indigenous microbiota of the lettuce. A cm-resistant 0157:H7 strain (H1827-cmR) was generated by inserting a nonpolar cmresistant cartridge at a noncoding region of the chromosome, using the method described by Datsenko and Wanner (14). Briefly, lambda Red recombinase expression plasmid pKD46 was electroporated into the lettuce outbreak 0157:H7 strain H1827. The cm-resistant cartridge insertion was confirmed by PCR, and DNA sequencing performed by ACGT Inc. (Wheeling, IL). The insertion of a cm-resistant cartridge into a noncoding region of the HI 827 chromosome did not alter the chlorine sensitivity of the 0157:H7 strain (data not shown).The antibiotic resistant non0157 STEC strains (026, 045, 0103, 0111, 0121, 0145, and 0104), designated as (strain#)-nalR, were constructed by subculturing through step-incremental nal concentrations (10, 20, and 50 pg/ml) in BHI agar plates at 37°C and were used in the transfer assays. Chlorine sensitivity assay. The free chlorine concentration in BHI was measured by using ChloroSense (Palintest Limited, Tyne and Wear, England), according to the manufacturer’s instruction. Bacterial growth in BHI containing a sublethal concentration of chlorine was examined as described (16). Briefly, a 1:10,000 dilution of a washed overnight culture, adjusted to 0.8 OD60o, was made in BHI broth containing 0 or 0.15 ppm of free chlorine. The free chlorine level was confirmed in the BHI immediately after chlorine addition. The cells were added immediately afterward. The growth of 200 pi of diluted culture was monitored in the Bioscreen C Automatic Microbiology Growth Curve System (Growth Curves USA, Piscataway, NJ) every 5 min for 24 h at 37°C. The lag phase of growth was defined as the time (hours) between initial inoculation and the time point when OD60o reached 0.1. The extension of the lag phase was defined as the difference between lag phases of the growth in the presence and absence of chlorine. All growth experiments were repeated at least three times for each strain. Transfer assays. Romaine lettuce was purchased from a local grocery store, and the three outer-most layers of leaves were removed. Transfer assays were conducted using the eight STEC

1862

DENG ET AL.

J. Food Prot., Vol. 77, No. 11

TABLE 1. Shiga toxigenic E. coli (STEC) strains used in this study Strain name

EDL933 Sakai HI 827 TW14359 DEC10B 97-3250 M T#10 TB352A MI03-19 MI01-88 MI05-14 DA-21 M T#80 TB154A 8419 PT91-24 RD8 3215-99 0201 9611 3007-85 MDCH-4 M T#2 M T#18 DA-5 DEC 101 4865/96 GS G5578620 IH 16 WS02568D WS03683D TY2482 B99BE001520 B9.9BE0011456 ATCC 35401

Serotype

0157:H7 0157:H7 0157:H7 0157:H7 026:H 11 026:H 11 026 0 2 6 :N 0 4 5 :H2 0 4 5 :H2 045:H2 0 4 5 :NM O103:H2 O103:H6 O103:H25 O103:N 011LH 2 0111:H8 OllLHll OllLNM 012LH19 012LH19 0121 012LH 19 0145:H16 0125:H28 0145:NM 0145:NT O104:H4 O104:H4 O104:H4 029:NM 073:H18 078:H 11

Place of isolation

United States Japan United States United States Australia (Brisbane) United States (Idaho) United States (Montana) United States (Washington) United States (Michigan) United States (Michigan) United States (Michigan) United States (Florida) United States (Montana) United States (Washington) United States (Idaho) United States (Washington) France United States (Texas) United States (Connecticut) United States (Nebraska) United States (Michigan) United States (Montana) United States (Montana) United States (Massachusetts) Canada Germany United States (Nebraska) Uruguay Egypt Egypt Germany United States (Texas) United States (Texas) United States

Reference/collection0

FDA FDA FDA FDA FDA

54 53 23 55 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 collection collection 9 collection collection collection

° FDA collection, U.S. Food and Drug Administration, 6502 South Archer Road, Bedford Park, IL. strains 0157:H7 H1827, 026 M T#10, 045:H2 M103-19, 0103: H6 TB154A, 0111:H2 RD8, 012LH 19 MDCH-4, 0145:NT IH16, and O104:H4 WS02568D, as well as the potential surrogates E. coli EcN and LAB. For each strain, a 2-ml aliquot of 18- to 24-h culture was washed in Butterfield’s phosphate buffer (BPB) three times by centrifugation (2,600 x g, 4°C for 10 min) and resuspended into 2 ml of BPB. The bacterial cell concentration was determined by serial dilution in BPB and plating on appropriate agars. As illustrated in Figure 1, pieces of lettuce lamina (leaf blade) were cut into a rectangular piece (2 by 4 cm), and 30 pi of bacterial suspension (~104 CFU/cm2) was evenly spread onto each piece. After drying in a biosafety cabinet for 1 h, the inoculated leaf was transferred into a 50-ml sterile conical tube containing uninocu­ lated lettuce (cut into narrow strips, comprising an equivalent surface area as the inoculated rectangular leaf) in 30 ml of either deionized water or 1 ppm of chlorine solution prechilled to 4°C. The lettuce pieces were washed by rotating the tube at 60 rpm for 1 min. After washing, the inoculated leaf was removed and transferred into another tube containing 5 ml of BPB and 0.5 ml of sodium thiosulfate (final concentration 0.1 M). Sodium thiosulfate (1 M and 1 ml) was then added to the washing tube to neutralize the chlorine in the wash water. The lettuce strips from the washing tube were transferred to another conical tube containing 5 ml of

BPB. The bacteria that were attached to the pieces were extracted by adding approximately five sterile glass beads (diameter, 6 mm; Fisher Scientific, Pittsburgh, PA) to the tubes and vortexing for 1 min. The cell population from the inoculated and washed leaf pieces, uninoculated strips, and wash water were enumerated by plating on BHI agar supplemented with appropriate antibiotics. Enumeration of the low numbers of bacterial cells that survived transfer in the presence of 1 ppm of chlorine was accomplished using membrane filtration, as follows: the BPB extract of the uninoculated strips (4 ml) or the wash water (29 ml) was filtered through Microfil membranes (Millipore, Billerica, MA), which were then incubated on selective agar plates. In addition, 1 ml of the wash water or the BPB extract for uninoculated strips was mixed with 1 ml of 2 x E. coli enrichment medium (Acumedia, Neogen, Lansing, MI) and incubated to confirm negative results obtained from the membrane filtration. A recovery control was performed by quantifying the inoculated bacteria (~ 104 total CFU/ cm2) on unwashed lettuce pieces by the bead-extraction method described previously. The transfer assay experiments for each strain were repeated at least twice. Chlorine inactivation assay. 0157:H7 strain H1827-cmR, 0 1 1 1:H2 strain RD8-nalR, and nonpathogenic EcN-nalR were used in the chlorine inactivation assay. For each strain, a culture was

STEC BEHAVIOR DURING LETTUCE WASHING

J. Food Prot., Vol. 77, No. 11

1863

Inoculated leaf and un­ inoculated leaves in 30 ml w a te r or 1 ppm chlorine

A fter washing at 60 rpm fo r 1 min, sort o u t the leaf and w ater samples for enum eration

V

Inoculated/ washed leaf in 5 ml BPB w ith 0.1 M Na2S20 3

'

i

Un­ inocu lated / washed leaves in 5 ml BPB

FIGURE 1. Experimental scheme o f lettuce wash fo r bacterial transfer assay. Inoculated bacteria are illustrated as the dots on the rectangular leaf piece in the figure. Uninoculated leaves are represented as narrow strips.

grown for 18 to 24 h, and 2 ml was washed in BPB three times and resuspended into 2 ml of BPB. Chlorinated water (~1 ppm of chlorine) was prepared, and the free and total chlorine concentra­ tions were determined on a Hach low-range chlorine meter (Hach, Loveland, CO). The chlorinated water was further diluted to 0.1 and 0.25 ppm in deionized water prechilled at 4°C. Bacterial suspension (—107 CFU/ml, 10 pi) was added to the 1 ml water, 0.1 or 0.25 ppm of chlorine, mixed for 5 s, and then 100 pi of 0.1 M sodium thiolsulfate was added to each for chlorine neutralization. The surviving bacteria were enumerated by plating 100 pi on appropriate selective agar plates (statistically relevant detection limit of this plating was —300 CFU/ml). The chlorine inactivation experiments were performed twice, with duplicate samples each time for each strain. Statistical analysis. The data analysis of the water wash results was performed using log values (log CFU per square centimeter for leaf samples and log CFU per milliliter for water samples), while the chlorine wash data were analyzed as CFU per square centimeter for leaf samples and CFU per milliliter for water samples. A general linear model was adopted for the comparisons among the strains, and Tukey’s test was used for multiple comparison corrections. All analyses were performed using SAS 9.2 for Windows (SAS Institute, Cary, NC), and statistical significance was defined as P < 0.05.

RESULTS AND DISCUSSION Sensitivity to sublethal concentration of chlorine during growth in broth. Chlorine levels can fall dram atically to below effective levels by w ashing process effects, such as high organic load or high pH; therefore, we perform ed a prelim inary com parison o f the STEC strains for their susceptibility to a sublethal concentration o f chlorine.

FIGURE 2. Chlorine sensitivity o f STEC strains and nonpathogenic E. coli strain EcN. Lag-phase extension was used to represent chlorine sensitivity during the growth in BHI containing 0.15 ppm o f free chlorine. The order o f the strains is the same as listed in Table 1. Other O serotypes are the 027, 073, and 078. The error bars stand fo r the standard deviations o f the average lag-phase extension from three biological repeats. * means no growth was detected within 24 h. The shaded bars represent the strains involved in the later transfer assays. The strains in the same serotype showing significant differences (P < 0.05) are marked with different letters.

An extension o f the lag phase o f the bacterial grow th curve has been used to represent sensitivity to antim icrobial agents or sanitizers in grow th m edia (36, 47, 51). In the current study, chlorine sensitivity w as exam ined by m easuring the extension o f the lag phase o f bacterial grow th in a m edium containing —0.15 ppm o f free chlorine. As show n in Figure 2, the m ajority o f the strains exhibited a 5- to 10-h grow th delay w hen chlorine was present in the m edium . Three strains show ed the highest sensitivity as dem onstrated by their inability to grow at all w ithin 24 h. For the 0 1 5 7 :H 7 strains, ED L933 (the first bar in the 0 1 5 7 group in Fig. 2), isolated from a ham burger-associated outbreak, and the spinach-associated outbreak strain T W 14359 (the fourth bar) show ed sim ilar chlorine sensitivity. T hey w ere not as susceptible to chlorine as strains Sakai (the second bar) and H I 827 (the third bar), w hich were involved in sprout and lettuce outbreaks, respectively. In som e serotypes, one strain out o f the four show ed dram atically distinctive chlorine sensitivity from the others in the same group. F or example, the 0 2 6 :H 1 1 isolate D EC10B (the first bar in the 0 2 6 group) was very sensitive to chlorine com pared w ith the other 0 2 6 strains. T he nonpathogenic E. coli EcN dem onstrated a longer lag-phase extension in the presence o f chlorine (m eaning greater chlorine sensitivity) than the five 0 1 5 7 :H 7 strains.

Bacterial transfer during lettuce washing. W e were interested in studying w hether the various strains behaved differently during the lettuce w ashing process. For these experim ents, w e selected one strain show ing relatively low er chlorine sensitivity from each n o n -0 1 5 7 group, along with the 0 1 5 7 :H 7 strain involved in a lettuce-associated outbreak (the shaded bars in Fig. 2), and w e exam ined their transfer from contam inated leaves to the w ash w ater and to uninoculated leaves.

1864

DENG ET AL.

(A) Transferred to water

J. Food Prot., Vol. 77, No. 11

(A)

(B)

(B) Transferred to clean leaves

FIGURE 4. Bacterial transfer from inoculated lettuce after lettuce washing in 1 ppm o f chlorine. The dot plot illustrates the number o f bacteria (total CFU) detected in wash water (A) and recovered from the uninoculated leaves (B) fo r each replicate. Six replicates were included fo r each strain; symbols represent different replicates within each serotype.

FIGURE 3. Bacterial transfer from inoculated lettuce after lettuce washing in water (no chlorine). (A) The population recovered from wash water (log CFU per milliliter). (B) The population recovered from the uninoculated leaves (log CFU per square centimeter). The box plots were summarized from at least six replicates fo r each strain. The line inside a box represents the median o f the data set. The upper and lower bars represent the highest and lowest values to a distance o f at most 1.5 interquartile ranges fo r each set. The circles represent outliers.

After washing in water (no chlorine), approximately 1 log CFU/cm2 of bacteria was removed from the inoculated leaf pieces, as reported in other studies (29, 38). Approximately 3 to 4 log CFU/ml was detected in the water sample (Fig. 3A). Similar numbers of bacteria were transferred into the wash water for all of the STEC strains. Pathogen transfer to clean uninoculated leaves was about 2 log CFU/cnr (Fig. 3B), with the 026 strain showing significantly less transfer than 0157:H7 (P < 0.05). No differences were demonstrated between the nonpathogenic EcN, LAB, and the 0157:H7 strain. In commercial practice, the chlorine concentration used in washing lettuce is approximately 100 to 150 ppm of chlorine (44), and the minimum level of chlorine sufficient for preventing cross-contamination during lettuce washing has been reported to be in the range of 5 to 10 ppm (21, 32); however, we could not use these lethal concentrations in our

transfer experiments because we would not be able to compare the strains for their relative survivability. For this reason, we chose the sublethal concentration 1 ppm to compare the various strains for their ability to transfer in the presence of chlorine. Fewer than 10 CFU were detected in the 1-ppm chlorine wash water for the majority of the strains (Fig. 4A); however, the 0 111 strain and EcN showed significantly more transfer to the 1 ppm of chlorine wash water as compared with 0157:H7 (P < 0.001), although the transfer was not consistent among replicates. Although the transfer characteristics demonstrated by this particular Ol 11 strain do not necessarily represent the behavior of other 0111 strains, reports in the literature have noted that certain 0111 strains are more acid resistant (6) or can form biofilms to be more tolerant to disinfectant treatment than 0157:H7 (50). The LAB strain did not show any significant difference compared with 0157 in the 1 ppm of chlorine wash. In our experiments on bacterial transfer to clean uninoculated leaves (Fig. 4B), we found that, except for EcN, the number of bacterial cells transferred by the STEC groups, as well as the LAB strain, was nearly undetectable. This result is in contrast to the transfer experiments reported by Zhang et al. (56), in which transfer to wash water or uninoculated leaves in the presence of 30 and 50 ppm was detectable. The different experimental procedures, including the methods for measuring chlorine, are potential reasons for the discrepant results. Sensitivity to sublethal concentration of chlorine in water. The relatively greater transfer to the 1 ppm of chlorinated water shown by 0111 and EcN, as well as relatively greater transfer to uninoculated leaves by EcN, imply that the ability to cross-contaminate might be higher

STEC BEHAVIOR DURING LETTUCE WASHING

J. Food Prat., Vol. 77, No. 11

4 £

3.5 -

D u, o bo O 2.5 o e o ■o.p •3

1.5

ID

c4

1



bo c J 0.5 0157

Olll

EcN

FIGURE 5. Bacterial survival in 0.25 ppm of chlorine solution. The log reduction was determined by comparing the remaining bacterial population after being added into 0.25 and 0 ppm of chlorine water. The error bars represent standard deviations of four experimental repeats.

for these strains than the others. However, EcN showed greater chlorine sensitivity compared with 0157 in the growth experiments (shown in Fig. 2), indicating that a correlation between chlorine sensitivity in broth growth and the extent of transfer during lettuce washing does not exist. The use of the lag-phase extension assay to assess chlorine sensitivity in actuality reflects a time course of complex bacterial responses to the presence of chlorine, including injury/inactivation, adjustment of the cells under oxidative stress, and then revival of the injured cells, all of which occurred over several hours in a nutrient medium. In contrast, the transfer assay measured the ability of cells to survive a 1-min chlorine exposure without nutrients and encompassed the total effects of both leaf detachment and chlorine sensitivity in solution. We, therefore, decided to compare sensitivity of 0157, 0111, and EcN to chlorine in solution, without leaf detachment, to separate the two factors. When the three strains were exposed to different concentrations of chlorinated water for 5 s, no significant log reduction was observed for any of them in 0.1 ppm of chlorine; however, in 0.25 ppm of chlorine, the 0157 strain showed significantly more log reduction than 0111 and EcN (Fig. 5). The results of chlorine inactivation correlated with the extent of transfer for these strains, i.e., with the order of sensitivity to chlorine being 0157 > O l l l > EcN and the extent of detachment from leaves following in the same order. EcN exhibited similar growth kinetics (data not shown) and higher sensitivity to the sublethal chlorine in BHI growth media compared with the 0157:H7 strains and no significant difference compared with 0157:H7 in the transfer assay without chlorine. However, EcN showed a greater extent of cross-contamination to water and uninoc­ ulated leaves in the presence of 1 ppm of chlorine and less susceptibility as compared with 0157:H7, when suspended in water containing sublethal chlorine (0.25 ppm). Consid­ ering the traits evaluated in this study, neither the EcN nor

1865

the LAB strain showed optimal characteristics as a surrogate for the STECs. Beyond these preliminary tests, more extensive evaluations need to be performed for surrogate evaluations (35, 42). For example, wash water obtained from a lettuce processing plant could be used to provide additional realistic conditions for evaluation, with results verified in a scaled-up washing system (15,31), and a safety assessment should be made (27). Nevertheless, the compar­ ison of traits as performed in this study may serve as a starting point for selecting surrogates that could be used in validating antimicrobial washing processes. In summary, our investigation evaluated the range of behavior of STEC with respect to sensitivity to sublethal levels of chlorine, detachment from inoculated leaves, and transfer to wash water and uninoculated leaves. These assays have targeted the phenotypic traits assumed to be relevant to pathogen behavior during lettuce washing processes and revealed a range of behaviors shown by the STEC strains. In the presence of 1 ppm of chlorine, transfer to the wash water and, in some cases, to clean uninoculated leaves was demonstrated for certain strains. Although 5 to 10 times below the minimum level of chlorine sufficient for preventing cross-contamination, as reported in other studies (21, 32), it is not inconceivable that such a low level could result in leaf microenvironments during lettuce washing due to fluctuations in organic load and other processing effects (3). This information supports the inclusion of the different STEC serotypes in evaluating preventive measures for 0157:H7 to ensure food safety. These assays and data may be useful to eventually develop a strategy by which potential surrogates for the pathogens can be identified for use in validating antimicrobial washing processes for lettuce or other fresh produce items. ACKNOWLEDGMENTS This study was supported by a FDA cooperative agreement grant to the Institute of Food Safety and Health of the Illinois Institute of Technology. Kaiping Deng was sponsored by the Oak Ridge Institute for Science and Education Research Participation Program through a contract with the FDA. The authors thank Dr. Tong-Jen Fu for valuable discussions related to fresh produce washing processes.

REFERENCES 1.

Adams, M. R„ A. D. Hartley, and L. J. Cox. 1989. Factors affecting the efficacy of washing procedures used in the production of prepared salads. Food Microbiol. 6:69-77. 2. Allende, A., M. V. Selma, F. Lopez-Galvez, R. Villaescusa, and M. I. Gil. 2008. Role of commercial sanitizers and washing systems on epiphytic microorganisms and sensory quality of fresh-cut escarole and lettuce. Postharvest Biol. Technol. 49:155-163. 3. Barrera, M. J., R. Blenkinsop, and K. Warriner. 2012. The effect of different processing parameters on the efficacy of commercial post­ harvest washing of minimally processed spinach and shredded lettuce. Food Control 25:745-751. 4. Beuchat, L. R. 1999. Survival of enterohemorrhagic Escherichia coli 0157:H7 in bovine feces applied to lettuce and the effectiveness of chlorinated water as a disinfectant. J. Food Prot. 62:845-849. 5. Brooks, J. T„ E. G. Sowers, J. G. Wells, K. D. Greene, P. M. Griffin, R. M. Hoekstra, and N. A. Strockbine. 2005. Non-0157 Shiga toxinproducing Escherichia coli infections in the United States, 19832002. J. Infect. Dis. 192:1422-1429. 6. Buchanan, R. L., and S. G. Edelson. 1996. Culturing enterohemor­ rhagic Escherichia coli in the presence and absence of glucose as a

1866

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

DENG ET AL.

simple means of evaluating the acid tolerance of stationary-phase cells. Appl. Environ. Microbiol. 62:4009-4013. Buchholz, A. L., G. R. Davidson, B. P. Marks, E. C. Todd, and E. T. Ryser. 2012. Quantitative transfer of Escherichia coli 0157:H7 to equipment during small-scale production of fresh-cut leafy greens. /. Food Prot. 75:1184-1197. Buchholz, A. L., G. R. Davidson, B. P. Marks, E. C. D. Todd, and E. T. Ryser. 2012. Transfer of Escherichia coli 0157:H7 from equipment surfaces to fresh-cut leafy greens during processing in a model pilot plant production line with sanitizer-free water. J. Food Prot. 75:1920-1929. Buchholz, U., H. Bernard, D. Werber, M. M. Bohmer, C. Remschmidt, H. Wilking, Y. Delere, M. an der Heiden, C. Adlhoch, J. Dreesman, J. Ehlers, S. Ethelberg, M. Faber, C. Frank, G. Fricke, M. Greiner, M. Hohle, S. Ivarsson, U. Jark, M. Kirchner, J. Koch, G. Krause, P. Luber, B. Rosner, K. Stark, and M. Kuhne. 2011. German outbreak of Escherichia coli O104:H4 associated with sprouts. N. Engl. J. Med. 365:1763-1770. Busta, F. F., T. V. Suslow, M. E. Paris, L. R. Beuchat, J. N. Farber, E. H. Garrett, and L. J. Harris. 2003. The use of indicators and surrogate microorganisms for the evaluation of pathogens in fresh and fresh-cut produce. Compr. Rev. Food Sci. Food Saf. 2(Suppl.):179185. Centers for Disease Control and Prevention. 2012. Multistate outbreak of Shiga toxin-producing Escherichia coli 026 infections linked to raw clover sprouts at Jimmy John’s restaurants. Available at: http://www.cdc.gov/ecoli/2012/O26-02-12/. Accessed 5 May 2014. Centers for Disease Control and Prevention. 2013. Surveillance for foodbome disease outbreaks—United States, 1998-2008. Morb. Mortal. Wkly. Rep. 62:1-35. Cutter, C. N., and M. Rivera-Betancourt. 2000. Interventions for the reduction of Salmonella Typhimurium D T I04 and non-0157:H7 enterohemorrhagic Escherichia coli on beef surfaces. J. Food Prot. 63:1326-1332. Datsenko, K. A., and B. L. Wanner. 2000. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. USA 97:6640-6645. Davidson, G. R., A. L. Buchholz, and E. T. Ryser. 2013. Efficacy of commercial produce sanitizers against nontoxigenic Escherichia coli 0157:H7 during processing of iceberg lettuce in a pilot-scale leafy green processing line. J. Food Prot. 76:1838-1845. Deng, K., S. Wang, X. Rui, W. Zhang, and M. L. Tortorello. 2011. Functional analysis of ycfR and ycfQ in Escherichia coli 0157:H7 linked to outbreaks of illness associated with fresh produce. Appl. Environ. Microbiol. 77:3952-3959. Food and Agriculture Organization of the United Nations and World Health Organization. 2008. Microbiological hazards in fresh leafy vegetables and herbs: meeting report. Microbiological Risk, Assess­ ment Series no. 14. Food and Agriculture Organization, Rome. Geomaras, I., H. Yang, G. Moschonas, M. C. Nunnelly, K. E. Belk, K. K. Nightingale, D. R. Woemer, G. C. Smith, and J. N. Sofos. 2012. Efficacy of chemical interventions against Escherichia coli 0157:H7 and multidrug-resistant and antibiotic-susceptible Salmonella on inoculated beef trimmings. J. Food Prot. 75:19601967. Gil, M. I., M. V. Selma, F. Lopez-Galvez, and A. Allende. 2009. Fresh-cut product sanitation and wash water disinfection: problems and solutions. Int. J. Food Microbiol. 134:37^15. Goodbum, C., and C. A. Wallace. 2013. The microbiological efficacy of decontamination methodologies for fresh produce: a review. Food Control 32:418—427. Gomez-Lopez, V. M., A.-S. Lannoo, M. I. Gil, and A. Allende. 2014. Minimum free chlorine residual level required for the inactivation of Escherichia coli 0157:H7 and trihalomethane generation during dynamic washing of fresh-cut spinach. Food Control 42:132-138. Gragg, S. E., and M. M. Brashears. 2010. Reduction of Escherichia coli in fresh spinach, using lactic acid bacteria and chlorine as a multihurdle intervention. J. Food Prot. 73:358-361. Hilbom, E. D., J. H. Mermin, P. A. Mshar, J. L. Hadler, A. Voetsch, C. Wojtkinski, M. Swartz, R. Mshar, M. A. Lambert-Fair, J. A.

J. Food Prot., Vol. 77, No. 11

24.

Farrar, M. K. Glynn, and L. Slutsker. 1999. A multistate outbreak of Escherichia coli 0157:H7 infections associated with consumption of mesclun lettuce. Arch. Intern. Med. 159:1758-1764. Kalchayanand, N., T. M. Arthur, J. M. Bosilevac, J. W. Schmidt, R. Wang, S. D. Shackelford, and T. L. Wheeler. 2012. Evaluation of commonly used antimicrobial interventions for fresh beef inoculated with Shiga toxin-producing Escherichia coli serotypes 026, 045, 0103, 0111, 0121, 0145 and 0157:H7. J. Food Prot. 75:1207-

1212. 25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Keskinen, L. A., A. Burke, and B. A. Annous. 2009. Efficacy of chlorine, acidic electrolyzed water and aqueous chlorine dioxide solutions to decontaminate Escherichia coli 0157:H7 from lettuce leaves. Int. J. Food Microbiol. 132:134-140. Kim, J. K., and M. A. Harrison. 2009. Surrogate selection for Escherichia coli 0157:H7 based on cryotolerance and attachment to romaine lettuce. J. Food Prot. 72:1385-1391. Kopit, L. M., E. B. Kim, R. J. Siezen, L. J. Harris, and M. L. Marco. 2014. Safety of the surrogate microorganism Enterococcus faecium NRRL B-2354 for use in thermal process validation. Appl. Environ. Microbiol 80:1899-1909. Kuda, T., T. Yano, and M. T. Kuda. 2008. Resistances to benzalkonium chloride of bacteria dried with food elements on stainless steel surfaces. LWT Food Sci. Technol. 41:988-993. Lang, M. M., L. J. Harris, and L. R. Beuchat. 2004. Survival and recovery of Escherichia coli 0157:H7, Salmonella, 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. 67:1092-1103. Lopez-Galvez, F., A. Allendem, M. V. Selma, and M. I. Gil. 2009. Prevention of Escherichia coli cross-contamination by different commercial sanitizers during washing of fresh-cut lettuce. Int. J. Food Microbiol. 133:167-171. Luo, Y., X. Nou, P. Millner, B. Zhou, C. Shen, Y. Yang, Y. Wu, Q. Wang, H. Feng, and D. Shelton. 2012. A pilot plant scale evaluation of a new process aid for enhancing chlorine efficacy against pathogen survival and cross-contamination during produce wash. Int. J. Food Microbiol. 158:133-139. Luo, Y., X. Nou, Y. Yang, I. Alegre, E. Turner, H. Feng, M. Abadias, and W. Conway. 2011. Determination of free chlorine concentrations needed to prevent Escherichia coli 0157:H7 cross-contamination during fresh-cut produce wash. J. Food Prot. 74:352-358. Malchow, H. A. 1997. Crohn’s disease and Escherichia coli. A new approach in therapy to maintain remission of colonic Crohn’s disease? J. Clin. Gastroenterol. 25:653-658. Michigan State University. 2006. STEC Center. Michigan State University. Available at: http://www.shigatox.net/stec/cgi-bin/non0157. Accessed 3 March 2014. National Advisory Committee on Microbiological Criteria for Foods. 2010. Parameters for determining inoculated pack/challenge study protocols. J. Food Prot. 73:140-202. Njume, C., A. J. Afolayan, A. M. Clarke, and R. N. Ndip. 2011. Crude ethanolic extracts of Garcinia kola seeds Heckel (Guttiferae) prolong the lag phase of Helicobacter pylori: inhibitory and bactericidal potential. J. Med. Food 14:822-827. Nou, X., and Y. Luo. 2010. Whole-leaf wash improves chlorine efficacy for microbial reduction and prevents pathogen cross­ contamination during fresh-cut lettuce processing. J. Food Sci. 75: M283-M290. Nou, X., Y. Luo, L. Hollar, Y. Yang, H. Feng, P. Millner, and D. Shelton. 2011. Chlorine stabilizer T-128 enhances efficacy of chlorine against cross-contamination by E. coli 0157:H7 and Salmonella in fresh-cut lettuce processing. J. Food Sci. 76:M218224. Olaimat, A. N., and R. A. Holley. 2012. Factors influencing the microbial safety of fresh produce: a review. Food Microbiol. 32:1— 19. Parish, M. E., L. R. Beuchat, T. V. Suslow, L. J. Harris, E. H. Garrett, J. N. Farber, and F. F. Busta. 2003. Methods to reduce/eliminate pathogens from fresh and fresh-cut produce. Compr. Rev. Food Sci. Food Saf. 2(Suppl): 161-173.

J. Food Prot., Vol. 77, No. 11

41.

Shen, C., Y. Luo, X. Nou, Q. Wang, and P. Millner. 2013. Dynamic effects of free chlorine concentration, organic load, and exposure time on the inactivation of Salmonella, Escherichia coli 0157:H7, and non-0157 Shiga toxin-producing E. coli. J. Food Prot. 76:386-393. 42. Sinclair, R. G., J. B. Rose, S. A. Hashsham, C. P. Gerba, and C. N. Haas. 2012. Criteria for selection of surrogates used to study the fate and control of pathogens in the environment. Appl. Environ. Microbiol. 78:1969-1977. 43. Skaloud, J„ L. Pokludova, R. Novotna, and A. Cizek. 2003. Evaluation by conductance assay of Shiga toxin-producing Esche­ richia coli (STEC) 0157 and 026 and their sensitivity to selected disinfectants. Acta Vet. Brno 72:101-109. 44. Suslow, T. V. 1997. Postharvest chlorination. Basic properties and key points for effective disinfection. Publication 8003. University of California, Division of Agriculture and Natural Resources. Available at: http://anrcatalog.ucdavis.edu/pdf/8003.pdf. Accessed 3 March 2014. 45. Taylor, E. V., T. A. Nguyen, K. D. Machesky, E. Koch, M. J. Sotir, S. B. Bohm, J. P. Folster, R. Bokanyi, A. Kupper, S. A. Bidol, A. Emanuel, K. D. Arends, S. A. Johnson, J. Dunn, S. Stroika, M. K. Patel, and I. Williams. 2013. Multistate outbreak of Escherichia coli 0145 infections associated with romaine lettuce consumption, 2010. J. Food Prot. 76:939-944. 46. U.S. Food and Drug Administration. 2013. Current good manufac­ turing practice and hazard analysis and risk-based preventive controls for human food. Proposed rule. Fed. Regist. 78:3646-3824. 47. Valero, M., and M. C. Salmeron. 2003. Antibacterial activity of 11 essential oils against Bacillus cereus in tyndallized carrot broth. Int. J. Food Microbiol. 85:73-81. 48. Van Haute, S., I. Sampers, K. Holvoet, and M. Uyttendaele. 2013. Physicochemical quality and chemical safety of chlorine as a reconditioning agent and wash water disinfectant for fresh-cut lettuce washing. Appl. Environ. Microbiol. 79:2850-2861.

STEC BEHAVIOR DURING LETTUCE WASHING

49.

1867

Virtro, R., P. Manas, 1. Alvarez, S. Condon, and J. Raso. 2005. Membrane damage and microbial inactivation by chlorine in the absence and presence of a chlorine-demanding substrate. Appl. Environ Microbiol. 71:5022-5028. 50. Wang, R., J. L. Bono, N. Kalchayanand, S. Shackelford, and D. M. Harhay. 2012. Biofilm formation by Shiga toxin-producing Esche­ richia coli 0157:H7 and non-0157 strains and their tolerance to sanitizers commonly used in the food processing environment. J. Food Prot. 75:1418-1428. 51. Wang, S., K. Deng, S. Zaremba, X. Deng, C. Lin, Q. Wang, M. L. Tortorello, and W. Zhang. 2009. Transcriptomic response of Escherichia coli 0157:H7 to oxidative stress. Appl. Environ. Microbiol. 75:6110— 6123. 52. Warriner, K., A. Huber, A. Namvar, W. Fan, and K. Dunfield. 2009. Recent advances in the microbial safety of fresh fruits and vegetables. Adv. Food Nutr. Res. 57:155-208. 53. Watanabe, H., A. Wada, Y. Inagaki, K. Itoh and K. Tamura. 1996. Outbreaks of enterohaemorrhagic Escherichia coli 0157:H7 infection by two different genotype strains in Japan, 1996. Lancet 348:831832. 54. Wells, J. G., B. R. Davis, I. K. Wachsmuth, L. W. Riley, R. S. Remis, R. Sokolow, and G. K. Morris. 1983. Laboratory investigation of hemorrhagic colitis outbreaks associated with a rare Escherichia coli serotype. J. Clin. Microbiol. 18:512-520. 55. Wendel, A. M., D. H. Johnson, U. Sharapov, J. Grant, J. R. Archer, T. Monson, C. Koschmann, and J. P. Davis. 2009. Multistate outbreak of Escherichia coli 0157:H7 infection associated with consumption of packaged spinach, August-September, 2006: the Wisconsin investigation. Clin. Infect. Dis. 48:1079-1086. 56. Zhang, G., L. Ma, V. H. Phelan, and M. P. Doyle. 2009. Efficacy of antimicrobial agents in lettuce leaf processing water for control of Escherichia coli 0157:H7. J. Food Prot. 72:1392-1397.

Copyright of Journal of Food Protection is the property of Allen Press Publishing Services Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Behavior of Shiga toxigenic Escherichia coli relevant to lettuce washing processes and consideration of factors for evaluating washing process surrogates.

Postharvest processes for fresh produce commonly include washing in water containing antimicrobial chemicals, such as chlorine; however, if the antimi...
5MB Sizes 0 Downloads 9 Views