Determination of 5-Log Reduction Times for Food Pathogens in Acidified Cucumbers during Storage at 10 and 25 C†‡

Outbreaks of acid-resistant foodborne pathogens in acid foods with pH values below 4.0, including apple cider and orange juice, have raised concerns about the safety of acidified vegetable products. For acidified vegetable products with pH values between 3.3 and 4.6, previous research has demonstrated that thermal treatments are needed to achieve a 5-log reduction in the numbers of Escherichia coli O157:H7, Listeria monocytogenes, or Salmonella enterica. For some acidified vegetable products with a pH of 3.3 or below, heat processing can result in unacceptable product quality. The purpose of this study was to determine the holding times needed to achieve a 5-log reduction in E. coli O157:H7, L. monocytogenes, and S. enterica strains in acidified vegetable products with acetic acid as the primary acidulant, a pH of 3.3 or below, and a minimum equilibrated temperature of 10 C. We found E. coli O157:H7 to be the most acid-resistant microorganism for the conditions tested, with a predicted time to achieve a 5-log reduction in cell numbers at 10 C of 5.7 days, compared with 2.1 days (51 h) for Salmonella or 0.5 days (11.2 h) for Listeria. At 25 C, the E. coli O157:H7 population achieved a 5-log reduction in 1.4 days (34.3 h). Acidified foods are defined as low-acid food products to which acid or acid food ingredients have been added. The final pH value for acidified foods must be at or below 4.6 for all ingredients. In the United States, regulations governing the safe manufacture of acidified vegetables were promulgated in 1979 (21 CFR part 114). The purpose of regulating these products, which have an excellent safety record, was to prevent botulism due to improper acidification. It has been shown that spore outgrowth and toxin production by Clostridium botulinum will not occur if the pH is maintained below 4.6 (9, 11). Recent outbreaks of disease caused by vegetative cells of acid-resistant food pathogens in some acid foods have caused concern about the safety of acidified vegetable products by the U.S. Food and Drug Administration and the acidified vegetable industry. Outbreaks of disease from Escherichia coli O157:H7 and Salmonella enterica strains have occurred in apple cider and orange juice (5, 6), which typically have pH values between 3.5 and 4.0. In response to these outbreaks, the juice hazard analysis critical control point regulation, 21 CFR part 120, was promulgated in 2001. This regulation mandates a 5-log reduction in acid-resistant bacterial pathogens, including E. coli O157:H7, which may be present * Author for correspondence. Tel: 919-513-0186; Fax: 919-513-0180; E-mail: [email protected]. † Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture or North Carolina Agricultural Research Service, nor does it imply approval to the exclusion of other products that may be suitable. ‡ Paper FSR07-03 of the journal series of the Department of Food Science, North Carolina State University, Raleigh. in juices. Because E. coli O157:H7 has been shown to be more heat and acid resistant than other bacterial food pathogens in juices (10), most research on the safety of juice products has focused on this organism. The U.S. Food and Drug Administration has raised concerns about the safety of acidified vegetable products that are not heat processed. The minimum times and temperatures for heat processing acidified cucumbers to ensure safety have been determined (3). However, acidified vegetable products with a pH of 3.3 or below may contain enough acid to ensure a 5-log reduction in bacterial pathogens without a heat treatment. These products typically contain 400 mM or greater acetic acid. Some of these products cannot be heat treated and maintain desirable sensory properties. Safety concerns by U.S. Food and Drug Administration and the acidified vegetable industry that E. coli O157:H7 could survive in acidified foods that did not receive a heat process are addressed by this study. In this article, we show that in a representative acidified vegetable product (cucumbers) with a pH at or below 3.3, a 5-log reduction in viable cell counts for acid-resistant pathogens occurred without a heat treatment within 6 days. We have shown that manufacturers of acidified foods with pH values at or below 3.3, with acetic acid as the primary acidulent, can therefore safely produce these products without heat treatment. Temperature has been shown to influence acid killing of bacterial pathogens under conditions typical of acidified vegetables (1). Refrigerated products were not considered in this study, however, and they are exempted from the J. Food Prot., Vol. 70, No. 11 SAFETY OF ACIDIFIED CUCUMBERS 2639 TABLE 1. Bacterial strainsa Strain ID Strain name Previous ID Origin B0195 Listeria monocytogenes SRCC 529 Pepperoni B0196 L. monocytogenes SRCC 1791 Yogurt B0197 L. monocytogenes SRCC 1506 Ice cream B0198 L. monocytogenes SRCC 1838 Cabbage B0199 L. monocytogenes SRCC 2075 Diced coleslaw B0200 Escherichia coli O157:H7 ATCC 43888 Human feces B0201 E. coli O157:H7 SRCC 1675 Apple cider outbreak B0202 E. coli O157:H7 SRCC 1486 Salami outbreak B0203 E. coli O157:H7 SRCC 2061 Ground beef B0204 E. coli O157:H7 SRCC 1941 Pork B0206 Salmonella Branderup SRCC 1093 10% salted yolk B0207 Salmonella Cerro SRCC 400 Cheese powder B0208 Salmonella Enteritidis SRCC 1434 Ice cream B0209 Salmonella Newport SRCC 551 Broccoli with cheese B0210 Salmonella Typhimurium SRCC 1846 Liquid egg a SRCC strains obtained from Silliker, Inc., Chicago, Ill.; ATCC, American Type Culture Collection, Manassas, Va.; ID, identification. acidified food regulations (21 CFR part 114). We therefore chose the most permissive conditions for survival of bacterial pathogens that are representative of nonpasteurized acidified vegetable products that have a pH at or below 3.3: 10 C, 400 mM acetic acid, and an ionic strength of 0.342 (equivalent to 2% NaCl, typical of many acidified vegetable products). Higher temperatures were not investigated because previous reports have shown that the acid sensitivity of E. coli increases with temperature (1). MATERIALS AND METHODS Preparation of pickle brines. Size 2B cucumbers were obtained a local supplier. Approximately 9 cucumbers (789 2 g) were packed in 571 2 ml of cover solution in 1.36-liter (46oz) glass jars to equilibrate at 2% sodium chloride, 0.1% calcium chloride, and pH 3.3. Vinegar (20% acetic acid) was used to achieve the target pH. After filling and sealing, the jars were heat processed so that the cold point of each jar was held at 74.4 C for 15 min. The jars were then cooled with cold tap water to a temperature of 40 C and allowed to reach ambient temperature of approximately 25 C in air. The metal jar lids had rubber septa (approximately 15 mm in diameter) inserted to allow aseptic sampling of the cover liquid without opening the jars or introducing contamination. The jars were stored at 25 C for at least 10 days to allow the equilibration of water-soluble components (e.g., acids, salt, sugars) between the cover solution and the cucumbers. Prior to use, the concentrations of organic acids and the pH of the brines in the jars were measured as described below. In some jars, pH values were adjusted aseptically by the addition of a small volume of HCl or NaOH. Jars were equilibrated at the indicated incubation temperatures (10 or 25 C) for a minimum of 48 h prior to inoculation, to allow temperature equilibration (data not shown), and then held at the incubation temperature for the duration of the experiment. Preparation and handling of bacterial cells. Bacteria used in this experiment consisted of cocktails of five strains for each of three different bacterial species (Table 1). For a given experiment, five strains of a single species were used. Bacteria were grown statically at 37 C for 16 h in tryptic soy broth (Difco, Becton Dickinson, Sparks, Md.) plus 1% glucose to induce acid resistance (4). Each of the five cultures was grown separately. Cells were harvested by centrifugation, concentrated by resuspending in 1/100 the original volume with sterile saline, and combined into a cocktail with approximately equal concentrations of each bacterial culture. The cell cocktails were added to the jars through the septum to give an initial total cell count of approximately 108 CFU/ml in each jar. Jar contents were mixed by manual inversion prior to sampling with approximately 1 ml of brine. At indicated time intervals, samples were removed and serially diluted with 10-fold dilutions prior to plating with a spiral plater (Spiral Biotech, Inc., Norwood, Mass.) on a neutral pH nonselective agar medium, tryptic soy agar (Difco, Becton Dickinson) plus 1% glucose. After 24 to 48 h of incubation at 37 C, colonies were counted with an automated spiral plate counter (Q-Count, Spiral Biotech). The lower detection limit was between 102 and 103 CFU/ml by this method. Biochemical analysis. Samples were withdrawn with a 1-ml syringe through the rubber septum in the jar lids for high-performance liquid chromatographic (HPLC) analysis to determine the concentrations of organic acids and the equilibrated pH. The pH was determined with an IQ240 pH meter (IQ Scientific Instruments, San Diego, Calif.). Organic acid concentrations were measured with a Thermo Separation Products HPLC (ThermoQuest, Inc., San Jose, Calif.) system consisting of a P2000 pump, an SCM100 solvent degasser, an AS3000 autosampler, and a UV6000 diode array detector (ThermoQuest). A Bio-Rad HPX-87H column, 300 by 7.8 mm (Bio-Rad Laboratories, Hercules, Calif.), was used to resolve malic, lactic, and acetic acids. The operating conditions of the system included a sample tray at 6 C, a column at 75 C, and 0.03 N H2SO4 eluent at a 1-ml/min flow rate. The UV6000 detector was set to 210 nm at a rate of 1 Hz for data collection. ChromQuest version 4.1 chromatography software was used to control the system and analyze the data, utilizing the peak heights for quantitative integration. Modeling and statistical analysis. Survival curves of CFU per milliliter versus time were generated and typically showed nonlinear killing kinetics. For this reason, a Weibull model was used for curve fitting to determine the 5-log reduction times as described by van Boekel (12): Predicted log survivors N0 [1/Ln(10)]( / )1/ Parameters of the model include the initial cell numbers (N0) and two shape parameters ( and ). The predicted survivor curve was plotted as log survivors versus time ( ). With parameters and , the 5-log reduction times were calculated: Predicted 5-log reduction time [ Ln(10 5)] Statistical analysis of the predicted 5-log reduction time versus temperature data was carried out by the NLIN procedure of the SAS program (SAS Institute Inc., Cary, N.C.) to determine the standard errors of the fitted data and the upper standard error for the 5-log reduction estimate (3). We added five times the standard error to the predicted 5-log reduction time, as was done previously for thermal treatments, to ensure at least a 5-log reduction of these pathogens (3). For each bacterial species tested, the data J. Food Prot., Vol. 70, No. 11 2640 BREIDT ET AL. FIGURE 1. The survival of E. coli O157:H7, S. enterica, and L. monocytogenes strains in acidified pickle jars at 10 C. The data for E. coli O157:H7 (circles), S. enterica (triangles), and L. monocytogenes (squares) show the log of the viable cell count from seven or more replicate experiments, each with a cocktail of five strains of a given species. The solid lines represent the predicted survival curves from the Weibull model. TABLE 2. The 5-log reduction times of E. coli O157:H7, Salmonella, and Listeria and parameters for Weibull curves Parametera Estimate SE of the estimate Predicted 5-log reduction timeb