Thermal Resistance of <I>Salmonella Enterica</I> Serotypes, <I>Listeria

703

Journal of Food Protection, Vol. 68, No. 4, 2005, Pages 703±710 Copyright ᮊ, International Association for Food Protection

Thermal Resistance of Salmonella enterica Serotypes, Listeria monocytogenes, and Staphylococcus aureus in High Solids Liquid Egg Mixes

X. LI,1 B. W. SHELDON,1* AND H. R. BALL2

1Department of Poultry Science, North Carolina State University, Raleigh, North Carolina 27695-7608; and 2Michael Foods Egg Products Company, 120 Tower Street South, Gaylord, Minnesota 55334, USA

MS 04-290: Received 29 June 2004/Accepted 8 December 2004 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021

ABSTRACT

Decimal reduction times (D-values) were determined for Salmonella enterica serotypes, Listeria monocytogenes, and ϭ ϭ Staphylococcus aureus in two high solids egg mixes designated A and B (water activity [aw] 0.76 and 0.82; solids 53.12 and 52.63%; pH ϭ 5.09 and 5.29; viscosity ϭ 183 and 119 centipoise/s, respectively) using a low-volume (0.06 ml) sealed glass capillary tube procedure. For Salmonella, D-values ranged from 0.035 (70ЊC) to 0.193 min (64ЊC) in product A and from 0.048 to 0.193 min in product B. For Listeria, D-values ranged from 0.133 (70ЊC) to 0.440 min (64ЊC) in product A and from 0.074 to 0.364 min in product B. For Staphylococcus, D-values ranged from 0.332 (70ЊC) to 1.304 min (64ЊC) in product A and from 0.428 to 1.768 min in product B. For Listeria, the D-values of all heating temperatures were signi®cantly higher (P Ͻ 0.01) in product A than in product B. The similar trend was also observed for Salmonella and Staphylococcus but only at 66ЊC for Salmonella and 64ЊC for Staphylococcus. Greater temperature dependence was observed for Salmonella inactivation in the low aw and low pH product (A), while the product (B) with the higher aw and pH had greater temperature dependence for Listeria. Compared across both egg mixes and all heating temperatures, the Staphylococcus strains were from 6.2 to 11.7 times more heat resistant than S. enterica serotypes and from 2.2 to 7.5 times more heat resistant than L. monocytogenes.

The U.S. egg industry has been striving to provide its pasteurization processes. Current USDA regulations stipu- customers with egg products that are convenient and safe. late that liquid, frozen, and dried whole egg, yolk, and To extend shelf life and save storage space, approximately white be pasteurized or otherwise treated to inactivate sal- 31% of all eggs produced in the United States are sent to monellae. USDA-mandated egg pasteurization speci®ca- commercial breaking operations for producing egg products tions listed in the code of federal regulations (Title 7, Sec- (1). Microbial contamination of table eggs and their prod- tion 59.570) require that every particle of egg be held for ucts, especially by salmonellae, has become an important at least a speci®ed time and temperature to assure complete public health concern. To address these safety issues, the pasteurization and to produce a Salmonella-negative prod- U.S. Department of Agriculture (USDA) (38) has pre- uct (2). The minimum time-temperature pasteurization scribed egg pasteurization standards that provide minimum curves prescribed by the USDA (37) are designed to yield process temperatures and holding times for different cate- a9-D process for Salmonella spp. in liquid whole egg and gories of processed egg products. Although pasteurization to assure ``approximately equal degrees of pasteurization requirements for most categories of egg products are cov- effectiveness'' in other egg products. Salmonella is the only ered under the USDA regulations (38), a number of new bacterial pathogen speci®cally addressed within the context processed egg products have emerged as growing segments of these regulations. of the total egg market. Because these new egg products Salmonella spp. have a long historical association with have varied compositions that may affect pasteurization contaminated eggs and egg products. Over the last decade, conditions, the existing egg pasteurization requirements by the egg-associated serotype Salmonella Enteritidis has product category do not necessarily apply to these new egg emerged as one of the leading causes of foodborne sal- products being developed by food companies. Prior to gain- monellosis in the United States, Canada, the United King- ing regulatory approval, all food companies intending to dom, and other nations (35). Besides Salmonella, other market new pasteurized egg products must ®rst provide the foodborne pathogens including Listeria monocytogenes and USDA with process validation test data including the ®nd- Staphylococcus aureus also constitute some degree of con- ings from their thermal inactivation studies. tamination risk in liquid egg products, and therefore it is The primary means of ensuring the microbiological safety of liquid egg products is to use appropriable egg prudent that investigators and companies evaluate the relative heat resistance of these two other pathogens in their * Author for correspondence. Tel: 919-515-5407; Fax: 919-515-7070; liquid egg products. L. monocytogenes has been previously E-mail: brian࿞[email protected]. isolated from commercially processed liquid whole egg in 704 LI ET AL. J. Food Prot., Vol. 68, No. 4

TABLE 1. Salmonella serotypes, Listeria monocytogenes, and Staphylococcus aureus strains used throughout this study Organisms Origin of isolates Source

Salmonella serotypes Enteritidis ATCC 4931 Clinical isolate ATCCa Typhimurium 2564 Egg layer house isolate B. W. Sheldonb Heidelberg ATCC 8326 Unknown ATCC Typhimurium ATCC 14028 Human isolate ATCC Listeria monocytogenes NCF-F1KK4 Raw liquid whole egg isolate P. M. Foegedingc NCF-U2K3 Raw liquid whole egg isolate P. M. Foegeding ATCC 19115 Human isolate ATCC ATCC 19111 Poultry isolate ATCC Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 Staphylococcus aureus ATCC 10390 Unknown ATCC ATCC 11371 Patient with fatal enteritis ATCC ATCC 6538 Human isolate ATCC ATCC 25923 Clinical isolate ATCC a American Type Culture Collection. b North Carolina State University. c P. M. Foegeding, deceased.

the United States (18) and Northern Ireland (26). Listeria MATERIALS AND METHODS poses a particular concern if present in extended shelf-life Determination of background micro¯ora, pH, aw, solids refrigerated egg products because of its ability to multiply content, and viscosity. Two proprietary high solids egg mixes in foods held at proper refrigerated storage temperatures containing yolk, albumin, and salt (designated products A and B) (10, 19). Moreover, it is one of the most heat-resistant veg- were provided in aseptic packages from a commercial source. The etative bacterial pathogens associated with foods (20). In egg mixes were formulated and given a thermal treatment with a raw liquid egg yolk, a ®ve-strain Listeria inoculum was ¯ow diversion pasteurization temperature of 64.4ЊC and holding found to be 4.8 to 6.7 times more heat resistant than Sal- time of 5 min. The products were subsequently packaged in Schol- monella enterica serotypes over the temperature range test- le bags containing 4.54 kg of egg mix. Approximately 4 kg of ed (32). each egg mix were then aseptically transferred to sterile laminate S. aureus is also an important foodborne pathogen. In cartons and shipped overnight to our laboratory. The background aerobic bacterial populations of each egg mix sample were esti- 1999, a S. aureus strain was isolated from a scrambled egg mated by surface plating 1 ml of each egg mix across three brain product that was incriminated in a foodborne disease out- heart infusion (BHI) agar plates (ca. 0.33 ml per plate; Becton break in Japan (24). S. aureus has also been isolated from Dickinson, Sparks, Md.), incubating at 37ЊC for 48 h, and then the surface of shell eggs (8, 25) and from the contents of enumerating the number of colonies on each plate and adding their broken eggs used in confectionery products manufactured totals together (detection limit of 1 organism per ml). pH values in Nigeria (15). It is of particular importance because of its of the egg mixes were measured at room temperature using a pH broad survival and growth requirements. With respect to meter (Accumet model 10, Fisher Scienti®c, Atlanta, Ga.) and Accumet combination electrode with Calomel reference. a levels water activity (aw), the staphylococci are unique in being w of duplicate 2.5-g samples per egg mix were determined at 25ЊC better able to grow at lower aw levels than any other non- halophilic bacteria (4). Besides a , S. aureus has a wide using an AquaLab Series 3 aw meter (Decagon, Pullman, Wash.). w Solids content was determined for duplicate 2.5-g samples per egg pH growth range of 4 to 9.8. Moreover, the organism pro- mix using the standard forced air drying oven method at 105ЊC duces heat-resistant enterotoxins that can withstand pas- for 24 h (Boekel model, Southern Scienti®c Co., Rochester, N.Y.). teurization processes (4). S. aureus is also of concern when Viscosity was measured for duplicate 20-g samples (at 25ЊC) of improper handling or temperature abuse occurs after pro- each egg mix using a StressTech controlled stress rheometer cessing. (ReoLogica Instruments AB, Lund, Sweden) with serrated cup and The objectives of this study were (i) to determine the bob. Viscosity was expressed as centipoise per second (cps). heat resistance (D- and zD-values) of S. enterica serotypes, L. monocytogenes, and S. aureus in two experimental high Inocula preparation. In this study four-strain cocktails of each bacterial pathogen (S. enterica serotypes, L. monocytogenes, solids liquid egg mixes using a low-volume (0.06 ml per and S. aureus) were prepared and used in the heat inactivation tube) immersed sealed glass capillary tube procedure and trials (see Table 1 for the description of the strain origin and (ii) to compare the derived thermal inactivation data from source). Stock cultures for each strain were maintained in double this study with current USDA minimum egg pasteurization strength BHI broth plus 20% (vol/vol) glycerol and stored at requirements for other high solids egg products and other Ϫ20ЊC. For each strain, an individual 10-ml BHI culture was ini- heating media. tially prepared (22 to 24 h, 37ЊC), and a 0.1-ml aliquot was trans- J. Food Prot., Vol. 68, No. 4 THERMAL INACTIVATION OF SALMONELLA, LISTERIA, AND STAPHYLOCOCCUS 705

a TABLE 2. pH, solids content, aw, and viscosity of two high solids egg mix products (products A and B) Solids content Viscosity b bbЊ b Egg mixes pH (%) aw (cps at 25 C)

A 5.09 Ϯ 0.09 53.12 Ϯ 0.12 0.76 Ϯ 0.02 183 Ϯ 2 B 5.29 Ϯ 0.07 52.63 Ϯ 0.12 0.82 Ϯ 0.04 119 Ϯ 1 a Values are means Ϯ standard deviations of two replicate trials (n ϭ 2). b Њ A 10% salted egg yolk product has pH, solids, aw, and viscosity values of 5.89, 50.42%, 0.866, and 300 to 500 cps (at 30 C), respectively (11). ferred to 20 ml of fresh BHI broth (22 to 24 h, 37ЊC). The bac- culated as the negative reciprocal of the slope of the linear re- terial cells were harvested by centrifugation (15 min at 9,000 ϫ gression lines. Mean log D-values versus temperatures were also Њ Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 g and 4 C), resuspended in 10 ml of sterile 0.1% peptone water, plotted for each pathogen, and zD-values (de®ned as the temper- pooled into a common tube (4 by 10 ml per pathogen), then cen- ature change required for the thermal destruction curve to traverse Ͻ ϭϪ Ϫ1 trifuged again. The cell pellets were held on ice ( 1 h) before one log cycle) were calculated, where zD slope . The k val- the inoculation of the liquid egg mixes. ue, or rate of inactivation at a tested temperature, was estimated as 2.303D. The activation energy of thermal inactivation (i.e., Ea Thermal inactivation and bacterial enumeration. The cell in kilojoules per mol) was derived from Arrhenius plots (i.e., neg- pellets were thoroughly dispersed in 5 ml of each egg mix. In- ative slope of the ln k in contrast to the inverse temperature [1/T] oculated liquid egg mixes (0.06 ml) were aseptically dispensed plot). into sterile glass capillary tubes (0.8- to 1.1-mm inside diameter by 90 mm long, no. 34502, Kimble, Vineland, N.J.) using a sterile Statistical analysis. Independent triplicate thermal inactiva- syringe ®tted with a 100-mm blunt stainless steel needle (Fisher tion trials for S. enterica serotypes and duplicate thermal inacti- Scienti®c). Filled tubes were held in an ice-water slurry while vation trials for L. monocytogenes and S. aureus were conducted completing the ®lling procedure (Ͻ20 min) and then ¯ame sealed. for each egg mix at four selected heating temperatures of 64, 66, Initial inoculum levels ranged from 9.3 to 9.8 log CFU/ml of the 68, and 70ЊC. The data were analyzed by the general linear model liquid egg mixes. The capillary tubes were placed upright in a procedure for analysis of variance. The general linear model in- mesh screen±covered test tube rack and completely immersed in cluded product type, temperature, and product by temperature in- a preheated circulating water bath equipped with a calibrated tem- teractions and the residual sources of variation used as the error perature control module accurate to Ϯ0.05ЊC (model DC1, Haake, term. When a signi®cant effect was observed, means were com- Inc., Karlsruhe, Germany). With a type K thermocouple probe pared using the Pdiff option of the SAS statistical analysis soft- linked to a digital thermometer (model 871, Omega Engineering ware program, version 6.04 (SAS Institute Inc., Cary, N.C.). Mod- Inc., Stamford, Conn.), come-up heating times over the tempera- el and parameter adequacy was considered signi®cant at P Յ 0.05 ture range tested were determined and found to range from 2 to unless otherwise noted. Each mean was the average of at least 4 s at the geometric center of each capillary tube. At six to eight two independent runs with duplicate tubes evaluated per sampling evenly spaced timing intervals, two tubes per pathogen were re- time. moved from the water bath, rapidly cooled by immersion in an ice-water slurry, and held for 5 min. The exterior surface of each RESULTS AND DISCUSSION capillary tube was sanitized for 1 min in a 200 ppm sodium hy- pochlorite solution and then rinsed four times in sterile distilled Background micro¯ora, pH, aw, solids content, and water. Individual capillary tubes were then aseptically transferred viscosity. The two experimental egg mixes (products A and to sterile plastic test tubes containing 6 ml of peptone water, B) had background aerobic bacterial populations of less crushed using a ¯ame-sterilized glass rod, and serially diluted in than 1.0 CFU/ml. The pH, solids content, aw, and viscosity sterile peptone water, and appropriate dilutions were plated in du- of the two mixes are summarized in Table 2. Although the plicate onto BHI agar plates using the spiral plate method (3). The two products contain both yolk and albumen, their physical detection limit of this procedure was 2.0 log CFU/ml. After in- and chemical properties were similar to a 10% salted egg Њ cubation for 48 h at 37 C, isolated colonies of each pathogen were yolk as described in the International Egg Pasteurization enumerated using the Protocol automatic colony counter system Manual (see Table 2) (11). Product A had lower pH, higher (software version 2.06.07, Synbiosis, Cambridge, UK). Salmonel- solids content, and higher viscosity values than product B. la was con®rmed using modi®ed lysine iron agar (Oxoid, Og- densburg, N.Y.), triple sugar iron (Oxoid) slants, and Salmonella As expected, product A, which contained the higher solids O antiserum poly A (Becton Dickinson). L. monocytogenes and content, had a lower aw. The intrinsic parameters of these S. aureus were con®rmed using Gram stain and catalase tests. two egg products (i.e., aw, solids content, pH) can greatly Initial populations of the four-strain cocktails dispersed in each in¯uence the heat-resistance characteristics of bacterial egg mix and sealed in capillary tubes were determined as de- pathogens (5, 12, 13, 33, 39). For instance, the heat resis- scribed above. tances of Salmonella, L. monocytogenes, Escherichia coli, S. aureus, and Streptococcus have been reported to increase Determination of D-, zD-, and Ea-values. Thermal inactivation experiments were conducted at 64, 66, 68, and 70ЊC for as aw in their environment decreases. Presumably, the each pathogen. Survivor curves (survivors [log CFU per milliliter] mechanism of inactivation at higher aw levels is related to versus heat inactivation time [min]) were plotted for each tem- protein denaturation that occurs in the cell at a higher rate perature. A best-®t linear regression line over the entire survivor when heated in an environment containing more available curve was derived for each temperature. D-values (min) were cal- water. Hurst et al. (14) suggested that increased bacterial 706 LI ET AL. J. Food Prot., Vol. 68, No. 4

TABLE 3. Mean D- and zD- values and activation energies (Ea) of heat inactivation of Salmonella serotypes in high solids egg mixes D-values (min) (SD) at selected temperaturesa zD-value Ea Egg mix 64ЊC66ЊC68ЊC70ЊC (ЊC) (R2)b (kJ/mol)

A 0.193 ac 0.142 A b 0.078 c 0.035 d 8.1 A 273 (0.001) (0.018) (0.004) (0.001) (0.95) (7.4±8.1) B 0.193 a 0.094 B b 0.074 c 0.048 d 10.4 B 213 (0.001) (0.002) (0.003) (0.001) (0.96) (9.9±11) a Mean of three replicate trials (n ϭ 3). b 95% con®dence intervals. c Means in columns with different small capital letters differ signi®cantly at P Ͻ 0.01; means in rows with different lowercase letters Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 differ signi®cantly at P Ͻ 0.05. heat resistance in the presence of salt is due, at least in part, ing the cells, a mechanism similar to that of drying (13). to an increased pool of amino acids and to greater external A third factor that can in¯uence bacterial heat resistance is osmotic pressures caused by the salt. Increased heat resis- the pH of the cell medium (21, 22, 34). For example, Pal- tance in the presence of salt may also be attributed to a umbo et al. (30) and Froning et al. (11) reported that Sal- reduction in aw in the environment immediately surround- monella is signi®cantly more heat resistant at lower albumen pH. Following a 3.5-min hold time at 56.6ЊC, they reported log reductions in Salmonella populations of 0.97 at pH 7.8, 1.64 at pH 8.2, 2.20 at pH 8.8, and 3.24 at pH 9.3. The two egg mix products evaluated in this study had low aw levels, had pH values below 5.3, and contained salt and lipid, each of which could have contributed to an increase in heat resistance of Salmonella. The mechanisms by which pH can affect the heat resistance of bacteria are not fully understood. Some studies have indicated that pH can affect bacterial cell membrane structure leading to a disruption of the cytoplasmic membrane (22). It has been suggested by Katsui et al. (17) and Beuchat and Worthing- ton (7) that differences in the ¯uidity of membrane lipids, largely a re¯ection of membrane lipid composition, may play a role in the heat resistance of bacteria. Because differences in bacterial test strains and intrinsic factors of the food systems being compared (e.g., pH, aw, and solids content) can affect bacterial heat resistance, it is necessary to keep these factors in mind when comparing the present study with previous published studies.

Thermal resistance of S. enterica serotypes in high solids egg mixes. Table 3 summarizes the D-values, zD- values, and Ea of Salmonella strains in egg mixes A and B at heating temperatures of 64, 66, 68, and 70ЊC. A survivor curve of S. enterica serotypes, L. monocytogenes, and S. aureus at a processing temperature of 70ЊC in product B is shown in Figure 1. Across both egg mixes and all four heating temperatures, R2 values (mean Ϯ standard devia- tion) averaged 0.94 Ϯ 0.01, 0.96 Ϯ 0.02, and 0.96 Ϯ 0.02 for S. enterica serotypes, L. monocytogenes, and S. aureus, respectively. D-values for Salmonella ranged from 0.035 (70ЊC) to 0.193 min (64ЊC) and from 0.048 (70ЊC) to 0.193 min (64ЊC) for products A and B, respectively. As expected, increases in heating temperatures signi®cantly reduced FIGURE 1. Reduction of Salmonella, Listeria monocytogenes, the D-values for both egg mixes. Moreover, S. enterica se- and Staphylococcus aureus in product B at a processing temper- rotypes were signi®cantly more heat resistant (51% in- ature of 70ЊC. crease) in product A than in product B at 66ЊC. As previ- J. Food Prot., Vol. 68, No. 4 THERMAL INACTIVATION OF SALMONELLA, LISTERIA, AND STAPHYLOCOCCUS 707

TABLE 4. Mean D- and zD-values and activation energies (Ea) of heat inactivation of Listeria monocytogenes in high solids egg mixes D-values (min) (SD) at selected temperaturesa zD-value Ea Egg mix 64ЊC66ЊC68ЊC70ЊC (ЊC) (R2)b (kJ/mol)

A 0.440 A ac 0.400 A b 0.198 A c 0.133 A d 10.8 A 205 (0.018) (0.004) (0.003) (0.001) (0.94) (9.5±12.0) B 0.364 B a 0.206 B b 0.116 B c 0.074 B d 8.6 B 258 (0.020) (0.014) (0.012) (0.002) (0.99) (8.1±9.0) a Mean of two replicate trials (n ϭ 2). b 95% con®dence intervals. c Means in columns with different small capital letters differ signi®cantly at P Ͻ 0.01; means in rows with different lowercase letters Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 differ signi®cantly at P Ͻ 0.05. ously documented, a reduction in egg white pH was shown monellae inactivation in the higher solids and lower pH to cause an increase in heat resistance of salmonellae (9, product. 11). Compared with product B, the lower pH and aw of According to the 1969 USDA Egg Pasteurization Man- product A may be partially responsible for the greater heat ual (37), the minimum pasteurization temperatures and resistance detected for the four Salmonella serotypes ob- times required for different egg products range from 56.7ЊC served in product A. for 3.5 min for albumen to 63.3ЊC for 3.5 min for salted In comparison with previous published studies, the D- whole egg, sugared yolk, and salted yolk. Based on D- values of the four-strain Salmonella inoculum in both prod- values generated in the present study, the amount of time uct A and B were similar to those found by Michalski et required to produce a 9-D process (i.e., 9-log reduction in Њ al. (23), who reported a D64ЊC-value of 0.21 min for Sal- salmonellae) would range from 0.315 (70 C) to 1.737 min monella Enteritidis in a 10% salted egg yolk product. More- (64ЊC) for product A and 0.432 (70ЊC) to 1.737 min (64ЊC) over, the International Egg Pasteurization Manual (11) re- for product B. Using the 1980 USDA minimum egg pas- ports D-values for a ®ve-strain Salmonella cocktail (two teurization holding time and temperature requirement of 3.5 Salmonella Enteritidis strains, one Salmonella Typhimu- min at 63.3ЊC (for salted whole egg) and the D-values cal- rium TM-1 strain, one Salmonella Blockley strain, and one culated for the 64ЊC process temperature, the estimated log Salmonella Heidelberg strain) of 0.90 min at 63.3ЊC, 0.41 reductions in salmonellae populations would be 18.1 for min at 65.6ЊC, 0.15 min at 67.8ЊC, and 0.06 min at 70.0ЊC products A and B. These estimates are based only on pas- in a 10% salted egg yolk product. In contrast, Palumbo et teurization holding times and do not re¯ect further popu- al. (29) reported much higher D-values of 11.50 min lation reductions that occur during the come-up and cool- (63.3ЊC), 6.44 min (64.4ЊC), 3.85 min (65.5ЊC), and 2.07 down portions of the process. As observed in the present min (66.7ЊC) in 10% salted egg yolk using a cocktail of study, the 1969 USDA minimum pasteurization require- Salmonella Enteritidis, Salmonella Typhimurium, and Sal- ments (3.5 min at 63.3ЊC for salted yolk) still provide a monella Senftenberg. Compared with the Palumbo et al. large margin of safety regarding the heat inactivation of (29) study where 9-ml glass tubes were used in the heat salmonellae. inactivation trials, the current study employed low-volume Thermal resistance of L. monocytogenes in high sol- glass capillary tubes as did the Michalski et al. (23) and ids egg mixes. Table 4 summarizes the D-values, zD-values, Froning et al. (11) studies. The lower D-values recorded in and Ea of L. monocytogenes strains in products A and B at each of the studies employing small volume capillary tubes processing temperatures of 64, 66, 68, and 70ЊC. D-values is attributed to the very rapid heating come-up times and ranged from 0.440 (64ЊC) to 0.133 min (70ЊC) for product more linear inactivation of the test strains. In contrast, stud- A and 0.364 (64ЊC) to 0.074 min (70ЊC) for product B. L. ies employing larger sample volumes contained in test tubes monocytogenes was signi®cantly more heat resistant across or ¯asks result in signi®cantly higher D-values (29). An- all process temperatures in product A (from 1.2 to 1.9 other potential factor contributing to the higher D-values times) than detected in product B. Regardless of the product recorded in Palumbo's paper is their choice of Salmonella type, the four-strain pool of L. monocytogenes was from test strains. Besides the Salmonella Enteritidis test strains, 1.5 to 3.8 times more heat resistant than the four-strain pool they also included Salmonella Senftenberg, one of the most of Salmonella. Similarly, Schuman and Sheldon (32) re- heat-resistant Salmonella serotypes, which is rarely found ported a 4.8 to 6.7 times greater heat resistance of a ®ve- associated with foodborne illness. Ng et al. (28) reported strain pool of L. monocytogenes in liquid egg yolk than a that Salmonella Senftenberg is about 30 times more heat ®ve-strain pool of Salmonella (60 to 62.2ЊC, small volume resistant than Salmonella Typhimurium. The smaller tem- [0.05-ml] glass capillary tube procedure). The differences perature dependent factor zD (and larger Ea) obtained for in the cell wall structures of these organisms, gram positive product A indicates greater temperature dependence for sal- versus gram negative, may play a role in their contrasting 708 LI ET AL. J. Food Prot., Vol. 68, No. 4

TABLE 5. Heat inactivation mean D-values for Listeria mono- min (64ЊC) for product B. Applying the USDA minimum cytogenes in selected liquid egg products pasteurization holding time requirement of 3.5 min (63.3ЊC

Temperature D-value zD-value for salted whole egg or yolk) and the D-values calculated (ЊC) (min) (ЊC) at 64ЊC, the estimated log reductions in L. monocytogenes populations would be 8.0 and 9.6 for products A and B, a Liquid whole egg 57 3.7 7.2 respectively. Schuman and Sheldon (32) reported a 4.6-D 60 1.8 process for a ®ve-strain pool of L. monocytogenes inoculum 63 0.55 Њ Raw liquid egg yolkb 60.0 1.34 6.1 in raw liquid egg yolk processed for 6.2 min at 60 C. The 61.1 0.89 data presented in this study provide a comparable margin 62.2 0.58 of safety for Listeria. Presently, there are no general rules Egg yolkc 61.1 1.41 6.8d as to what constitutes an adequate level of Listeria inacti- 63.3 0.81 vation in heat-processed, refrigerated foods. Pasteurization 64.4 0.44 parameters for speci®c egg products are generally devel- Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 c d Egg yolk plus 10% salt 63.3 10.5 5.4 oped based on actual Listeria spp. incidence, population 64.4 6.11 data from microbiological surveillance studies, the relative 65.5 4.30 growth potential risks during refrigerated storage, and the 66.7 2.39 effectiveness of the plant's sanitation program. Relatively a Data cited from Muriana et al. (27) using a capillary tube pro- few published incidence or population data are available on cedure. this pathogen in raw liquid whole egg and egg yolk. How- b Data cited from Schuman and Sheldon (32) using a capillary ever, Moore and Madden (26) estimated the populations of tube procedure. Listeria spp. in samples of raw liquid whole egg collected c Data cited from Palumbo et al. (29) using a 9-ml tube procedure. on nine consecutive days from a commercial processing d Calculated from reported D-values. plant. The mean and maximum Listeria populations were 1.0 and 40 CFU/ml, respectively. heat-resistance values. The smaller zD- and higher Ea-values obtained in product B indicates greater temperature depen- Thermal resistance of S. aureus in high solids egg dence for Listeria inactivation in the lower solids egg prod- mixes. Table 6 summarizes the D-values, zD-values, and Ea uct. of the four-strain inoculum of S. aureus suspended in the Table 5 summarizes heat-resistance data taken from two egg mixes and processed at 64, 66, 68, and 70ЊC. D- several previously published studies involving L. monocy- values ranged from 1.304 (64ЊC) to 0.332 min (70ЊC) for togenes suspended in different liquid egg products. The Lis- product A and 1.768 (64ЊC) to 0.428 min (70ЊC) for prod- Њ teria heat-resistance data recorded in the present study are uct B. The derived zD- and Ea-values were 10.8 C and 205 similar to the D-values reported by Muriana et al. (27) (0.55 kJ/mol for product A and 10.2ЊC and 216 kJ/mol for prod- min at 63ЊC in liquid whole egg) and Schuman and Sheldon uct B. In a previous study (36), the heat resistance of two (32) (0.58 min at 62.2ЊC in raw liquid egg yolk). In con- S. aureus ATCC strains was determined at 56ЊC using 160- trast, Palumbo et al. (29) reported much higher D-values of ml screw cap bottles containing either casein peptone soy- 10.5 min at 63.3ЊC, 6.11 min at 64.4ЊC, 4.30 min at 65.5ЊC, meal peptone plus 0.5% yeast extract or skim milk. Com- and 2.39 min at 66.7ЊC in a 10% salted egg yolk product pared with their data (D-values from 10.2 to 13.2 min in having an aw of 0.865. casein peptone soymeal peptone plus 0.5% yeast extract Based on the D-values calculated in the present study, and from 9.2 to 9.8 min in skim milk), the D-values cal- the amount of time required to produce a 9-D process (i.e., culated in this study were substantially smaller (5 to 40 9-log reduction in Listeria) would range from 1.20 (70ЊC) times), presumably because of the higher heating temper- to 3.96 min (64ЊC) for product A and 0.67 (70ЊC) to 3.28 atures (8 to 14ЊC higher) and much smaller heating vessels.

TABLE 6. Mean D- and zD-values and activation energies (Ea) of heat inactivation of Staphylococcus aureus in high solids egg mixes D-values (min) (SD) at selected temperaturesa zD-value Ea Egg mix 64ЊC66ЊC68ЊC70ЊC (ЊC) (R2)b (kJ/mol)

A 1.304 A ac 0.885 b 0.748 b 0.332 c 10.8 205 (0.052) (0.010) (0.052) (0.014) (0.92) (10.1±11.4) B 1.768 B a 1.108 b 0.867 b 0.428 c 10.2 216 (0.113) (0.138) (0.120) (0.019) (0.96) (10.0±10.5) a Mean of two replicate trials (n ϭ 2). b 95% con®dence intervals. c Means in columns with different small capital letters differ signi®cantly at P Ͻ 0.01; means in rows with different lowercase letters differ signi®cantly at P Ͻ 0.05. J. Food Prot., Vol. 68, No. 4 THERMAL INACTIVATION OF SALMONELLA, LISTERIA, AND STAPHYLOCOCCUS 709

Poland and Sheldon (31) reported a D-value of 1.22 min of Listeria monocytogenes in liquid whole egg. J. Food Prot. 53:9± (56ЊC) for S. aureus ATCC 12600 in peptone water, which 14. 11. Froning, G. W., D. Peters, P. Muriana, K. Eskridge, D. Travnicek, and is similar to our results. Based on the ®ndings of this study, S. S. Sumner. 2002. International egg pasteurization manual (prepared the low-volume glass capillary tube method was suitable in cooperation with the United Egg Association, Alpharetta, Ga., and for generating linear survivor curves without ``shoulder'' the American Egg Board, Park Ridge, Ill.). United Egg Association, and tailing effects. Alpharetta, Ga. Compared with Salmonella and L. monocytogenes, S. 12. Garibaldi, J. A., R. P. Straka, and K. Ijichi. 1969. Heat resistance of aureus was substantially more heat resistant than Salmo- Salmonella in various egg products. Appl. Microbiol. 17:491±496. 13. Ghazala, S., D. Coxworthy, and T. Alkanani. 1995. Thermal kinetics nella (6.2 to 11.7 times larger D-values) and L. monocy- of Streptococcus faecium in nutrient broth/sous vide products under togenes (2.2 to 7.5 times larger D-values). The greater heat pasteurization conditions. J. Food Process. Preserv. 19:243±257. resistance of S. aureus may be related to several factors, 14. Hurst, A., G. S. Hendry, A. Hughes, and B. Paley. 1976. Enumeration including its ability to grow at aw as low as 0.83 (16). of sublethally heated staphylococci in some dried foods. Can. J. Mi- Although it is not anticipated that this organism will grow crobiol. 22:677±683. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021 in the two high solids egg mixes (a of 0.76 and 0.82), the 15. Ibeh, I. N., and Y. S. Izuagbe. 1986. An analysis of the micro¯ora of w broken eggs used in confectionery products in Nigeria and the oc- organism may persist for an extended period of time. Based currence of enterotoxigenic gram-negative bacteria. Int. J. Food Mi- on the D-values generated in the present study, the amount crobiol. 3:71±77. of time required to produce a 9-D process would range 16. Jay, J. M. 2000. Modern food microbiology, 6th ed., p. 35±57. Aspen from 2.99 (70ЊC) to 11.74 min (64ЊC) for product A and Publications, Gaithersburg, Md. 3.85 (70ЊC) to 15.91 min (64ЊC) for product B. Applying 17. Katsui, N., T. Tsuchido, M. Takano, and I. Shibasaki. 1981. Effect of preincubation temperature on the heat resistance of Escherichia coli the USDA minimum pasteurization holding time require- Њ having different fatty acid compositions. J. Gen. Microbiol. 122:357± ment of 3.5 min at 63.3 C for salted whole egg to the D- 361. values calculated at 64ЊC in our study, the estimated pop- 18. Leasor, S. B., and P. M. Foegeding. 1989. Listeria species in com- ulation reductions would be log 2.7 and 2.0 for products A mercially broken raw liquid whole egg. J. Food Prot. 52:777±780. and B, respectively. The ®ndings from this study provide 19. Lovett, J. 1989. Listeria monocytogenes, p. 283±310. In M. P. Doyle laboratory heat inactivation reference data for developing (ed.), Foodborne bacterial pathogens. Marcel Dekker, Inc., New York. 20. Mackey, B. M., and N. Bratchell. 1989. The heat resistance of Listeria commercial pasteurization processes for egg products con- monocytogenes: a review. Lett. Appl. Microbiol. 9:89±94. taining a high solids content, relatively low aw, and acidic 21. ManÄas, P., R. PagaÂn, J. Raso, and S. CondoÂn. 2003. Predicting thermal pH. inactivation in media of different pH of Salmonella grown at different temperatures. Int. J. Food Microbiol. 87:45±53. ACKNOWLEDGMENTS 22. Mendonca, A. F., T. L. Amoroso, and S. J. Knabel. 1994. Destruction This research was funded in part by the North Carolina Agricultural of gram-negative foodborne pathogens by high pH involves disruption Research Service and Michael Foods Egg Products Company, 120 Tower of the cytoplasmic membrane. Appl. Environ. Microbiol. 60:4009± Street South, Gaylord, MN 55334. The authors thank Susan Hale, Ashley 4014. Tyner, and Jessica LeMieux for their excellent technical assistance. 23. Michalski, C. B., R. E. Brackett, Y. C. Hung, and G. O. I. Ezeike. 1999. Use of capillary tubes and plate heat exchanger to validate U.S. REFERENCES Department of Agriculture pasteurization protocols for elimination of Salmonella Enteritidis from liquid egg products. J. Food Prot. 62: 1. American Egg Board. 2003. Distribution of shell eggs. Available at: 112±117. http://www.aeb.org/eii/facts/shell-egg-dist.html. Accessed 18 April 24. Miwa, N., A. Kawamura, T. Masuda, and M. Akiyama. 2001. An 2004. outbreak of food poisoning due to egg yolk reaction-negative Staph- 2. Anonymous. 1995. Regulations governing the inspection of eggs and ylococcus aureus. Int. J. Food Microbiol. 64:361±366. egg products (7 CFR, Part 59, 1 January 1995). USDA Agricultural 25. Moats, W. A. 1980. Classi®cation of bacteria from commercial egg Marketing Service, Poultry Division, Washington, D.C. washers and washed and unwashed egg. Appl. Environ. Microbiol. 3. AOAC International. 1998. Aerobic plate count, chap. 3. In Bacteri- 40:710±714. ological analytical manual, 8th ed. AOAC, Gaithersburg, Md. 26. Moore, J., and R. H. Madden. 1993. Detection and incidence of Lis- 4. Bergdoll, M. S. 1989. Staphylococcus aureus, p. 463±523. In M. P. teria species in blended raw egg. J. Food Prot. 56:652±654, 660. Doyle (ed.), Foodborne bacterial pathogens. Marcel Dekker, Inc., New 27. Muriana, P. M., H. Hou, and R. K. Singh. 1995. A ¯ow-injection York. system for studying heat inactivation of Listeria monocytogenes and 5. Beuchat, L. R. 1975. Environmental factors affecting survival and Salmonella enteritidis in liquid whole egg. J. Food Prot. 59:121±126. growth of Vibrio parahaemolyticus. A review. J. Milk Food Technol. 28. Ng, H., H. G. Bayne, and J. A. Garibaldi. 1969. Heat resistance of 38:476±480. 6. Beuchat, L. R. 1978. Injury and repair of gram-negative bacteria, with Salmonella: the uniqueness of Salmonella senftenberg 775W. J. Food special consideration of the involvement of the cytoplasmic mem- Sci. 17:78±82. brane. Adv. Appl. Microbiol. 23:219±243. 29. Palumbo, M. S., S. M. Beers, S. Bhaduri, and S. A. Palumbo. 1995. 7. Beuchat, L. R., and R. E. Worthington. 1976. Relationship between Thermal resistance of Salmonella spp. and Listeria monocytogenes in heat resistance and phospholipid fatty acid composition of Vibrio par- liquid egg yolk and egg yolk products. J. Food Prot. 58:960±966. ahaemolyticus. Appl. Environ. Microbiol. 31:389±394. 30. Palumbo, M. S., S. M. Beers, S. Bhaduri, and S. A. Palumbo. 1996. 8. Board, R. G., J. C. Ayres, A. A. Kraft, and R. H. Forsythe. 1964. The Thermal resistance of Listeria monocytogenes and Salmonella spp. in microbiological contamination of egg shells and egg packing mate- liquid egg white. J. Food Prot. 59:1182±1186. rials. Poult. Sci. 43:584±594. 31. Poland, A. L., and B. W. Sheldon. 2001. Altering the thermal resis- 9. Elliot, R. P., and B. C. Hobbs. 1980. Eggs and egg products, chap. tance of foodborne bacterial pathogens with an eggshell membrane 19. In J. H. Silliker (ed.), Microbial ecology of foods, vol. II. Food waste by-product. J. Food Prot. 64:486±492. commodities. International Commission on Microbiological Speci®- 32. Schuman, J. D., and B. W. Sheldon. 1997. Thermal resistance of Sal- cations for Foods. Academic Press, New York. monella spp. and Listeria monocytogenes in liquid egg yolk and egg 10. Foegeding, P. M., and S. B. Leasor. 1990. Heat resistance and growth white. J. Food Prot. 60:634±638. 710 LI ET AL. J. Food Prot., Vol. 68, No. 4

33. Sumner, S. S., T. M. Sandros, M. C. Harmon, V. N. Scott, and D. T. Staphylococcus aureus resulting from anaerobic heating and enumer- Bernard. 1991. Heat resistance of Salmonella typhimurium and Lis- ation of survivors. Food Microbiol. 11:275±280. teria monocytogenes in sucrose solutions of various water activities. 37. U.S. Department of Agriculture. 1969. Egg pasteurization manual. J. Food Sci. 56:1741±1743. ARS 74-48. Poultry Laboratory, Agricultural Research Service, U.S. 34. Teo, Y. L., T. J. Raynor, K. R. Ellajosyula, and S. J. Knabel. 1996. Department of Agriculture, Albany, Calif. Synergistic effect of high temperature and high pH on the destruction 38. U.S. Department of Agriculture. 1980. Regulations governing inspec- of Salmonella enteritidis and Escherichia coli O157:H7. J. Food Prot. tion of egg products. 7 CFR, Part 2859. Effective 8 May 1980. 59:1023±1030. 39. Veeramuthu, G. J., J. F. Price, C. E. Davis, A. M. Booren, and D. M. 35. Todd, E. C. D. 1996. Worldwide surveillance of foodborne disease: Smith. 1998. Thermal inactivation of Escherichia coli O157:H7, Sal- the need to improve. J. Food Prot. 59:82±92. monella senftenberg, and enzymes with potential as time-temperature 36. Ugborogho, T. O., and S. C. Ingham. 1994. Increased D-values of indicators in ground turkey thigh meat. J. Food Prot. 61:171±175. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/68/4/703/1676202/0362-028x-68_4_703.pdf by guest on 30 September 2021