Effects of Sodium Citrate Plus Sodium Diacetate and Buffered Vinegar on Escherichia Coli O157:H7 and Psychrotrophic Bacteria in Brine-Injected Beef

359

Journal of Food Protection, Vol. 74, No. 3, 2011, Pages 359–364 doi:10.4315/0362-028X.JFP-10-294 Copyright G, International Association for Food Protection

Effects of Sodium Citrate plus Sodium Diacetate and Buffered Vinegar on Escherichia coli O157:H7 and Psychrotrophic Bacteria in Brine-Injected Beef

AMUDHAN PONRAJAN,1,2 MARK A. HARRISON,1 JACOB R. SEGERS,2 BRADLEY K. LOWE,2 RUSSELL O. MCKEITH,2 T. DEAN PRINGLE,2 KARINA G. MARTINO,1 JAKE H. MULLIGAN,1 AND ALEXANDER M. STELZLENI2* Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 1Department of Food Science and Technology and 2Department of Animal and Dairy Sciences, University of Georgia, Athens, Georgia 30602, USA

MS 10-294: Received 15 July 2010/Accepted 4 December 2010

ABSTRACT The objective of this research was to examine the effects of sodium citrate plus sodium diacetate or buffered vinegar on Escherichia coli O157:H7 and psychrotrophic bacteria when incorporated in brine solutions for injected beef. Two experiments were conducted in which 30 top rounds and 30 top sirloins were injected (110%) to contain (i) 0.5% sodium chloride and 0.4% sodium tripolyphosphate as the control (CNT); (ii) CNT with a 1% solution of 80% sodium citrate plus 20% sodium diacetate (SCzD); or (iii) CNT with 2% buffered vinegar (VIN) in the final product. For the E. coli challenge, muscles were surface inoculated to target 6 log CFU/cm2. After injection and 10 days of storage in a vacuum package (4uC), one half of each muscle was sampled raw and the other half was cooked to an internal temperature of 60uC with a 12-min hold. For raw samples, a significant reduction of 0.6 and 1.0 log CFU/g of E. coli O157:H7 was observed in both SCzD- and VIN-injected top rounds and sirloins, respectively. All cooked samples were E. coli O157:H7 negative. For psychrotrophic analysis, subprimals were injected and vacuum packaged for 10 days at 0 ¡ 1uC. After 10 days of storage, steaks were fabricated and placed in aerobic display (4 ¡ 1uC) for 1, 7, 14, and 21 days. Psychrotrophic organism growth was restricted in SCzD and VIN samples when compared with CNT on all days except day 1. Sodium citrate plus sodium diacetate or buffered vinegar may improve the safety and shelf life of multineedle brine-injected beef.

Escherichia coli O157:H7 was first identified as a process. Sporing (13) has shown that when subprimals were foodborne pathogen in 1982, when it was associated with contaminated with pathogens such as E. coli (K-12), the two outbreaks of hemorrhagic colitis caused by the pathogen could be internalized during multineedle injection. consumption of undercooked frozen ground beef patties Current research has focused on a two-step injection process (3, 7, 9). A 1993 multistate outbreak of E. coli O157:H7 to reduce or eliminate E. coli contamination, in which the associated with hamburgers at a national fast-food chain subprimal undergoes a surface decontamination step increased the awareness about this pathogen (2). In followed by multineedle injection (4, 6). Little research response, the U.S. Department of Agriculture, Food Safety has been conducted to streamline this process by examining Inspection Service (USDA-FSIS) declared E. coli O157:H7 the efficacy of including antimicrobials directly in the brine an adulterant in ground beef in October 1994 under the solution (11, 19, 20). Furthermore, thermal processing has Federal Meat Inspection Act (4, 16). In January 1999, the been proven to be an effective method of reducing or USDA-FSIS further clarified that E. coli O157:H7 was also eliminating internalized pathogens in whole muscle, non- considered an adulterant in nonintact and whole muscle, intact beef in experimental settings (10, 13). However, the nonintact beef (8, 15). However, three outbreaks of E. coli thermal destruction of internalized pathogens needs to be O157:H7 in August 2000, June 2003, and August 2004 evaluated under simulated commercial conditions for roast were associated with whole muscle, nonintact beef products. beef products. Therefore, the objectives of this research In 2005, the USDA-FSIS mandated that processors were to examine the efficacy of sodium citrate plus sodium producing whole muscle, nonintact beef products must diacetate or buffered vinegar in the brine solution (i) against reassess their hazard analysis and critical control point E. coli O157:H7 in beef top rounds and sirloins and (ii) on plans, specifically addressing the biological hazard E. coli psychrotrophic bacterial growth for beef steaks after O157:H7 (17). extended aerobic storage. E. coli O157:H7 adulteration of beef subprimals is MATERIALS AND METHODS primarily from surface contamination during the harvesting Experiment 1: E. coli O157:H7 cultures. Four strains of E. * Author for correspondence. Tel: 706-583-0398; Fax: 706-542-0399; coli O157:H7 (ATCC 43888, human fecal isolate; E0122, cattle E-mail: [email protected]. isolate; K3995, spinach isolate; and F4546, alfalfa sprout outbreak 360 PONRAJAN ET AL. J. Food Prot., Vol. 74, No. 3 isolate) were used for this study. E. coli O157:H7 strains were Sealed Air Corporation) vacuum packager and stored in incubators activated by three consecutive transfers from fresh stock cultures (LR5 TempGuard, MedRep, Inc., Newnan, GA) at 4 ¡ 2uC for (280uC) into 9 ml of tryptic soy broth (Difco, BD, Sparks, MD) 10 days to simulate transportation and storage. containing 50 mg/ml of ampicillin (ampicillin sodium salt, Sigma- Aldrich, St. Louis, MO) at 24 ¡ 2 h intervals incubated at 37 ¡ Experiment 1: subprimal sampling. After 10 days of 2uC. On the day of the experiment, 1 ml of the overnight culture storage, subprimals were removed from their vacuum package and from each strain was transferred into 9 ml of fresh tryptic soy broth placed in sterile stainless steel pans. The pH and weight of the plus ampicillin. Three-milliliter samples from each strain were samples were recorded. The subprimals were split into two halves, pooled together to form the initial inoculum cocktail (12 ml) with each top round half having one inoculation site (top round, 8 containing approximately 107 CFU/ml. by 8 cm; top sirloin, 6 by 6 cm). One half of each muscle was sampled raw; the other half was cooked to an internal temperature Experiment 1: meat procurement and inoculation. of 60uC (with a 12-min hold) following an eight-step commercial Institutional Meat Purchase Specifications 169A beef top rounds roast beef cycle using a commercial smokehouse (Alkar model (30) and 184B beef top sirloins (30) from cull cows were obtained 450, Alkar-RapidPak, Inc., Lodi, WI) and USDA-FSIS guidelines

(FPL Food, LLC, Augusta, GA) and transported (0 ¡ 2uC) to the (14) for a 6.5 to 7.0 log reduction of Salmonella spp. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 University of Georgia Meat Science and Technology Center Sampling was conducted by turning the subprimal upside (Athens). On the day of the experiment, 10 top rounds or top down and aseptically removing a thin layer (0.2 to 0.3 cm) from sirloins were placed in sterile stainless steel pans on sterile stainless the bottom surface. A sterile stainless steel block template (5 by steel wire racks inside a biological safety cabinet (Class II, Type 5 cm) was used to create an impression on the shaved surface. The A2, NuAire, Plymouth, MN). subprimal was then cut along the impression of the template using To calculate the inoculation level, overnight cultures were a sterile knife for each cut. The excised sample was placed on a enumerated to determine the number of E. coli O157:H7 colonies sterile cutting board and divided into three sections based on height for each strain, and then calculations were made to determine how (top, middle, and bottom). Each section was aseptically separated many dilutions were needed and how many microliters of the from the sample block. Sample sections were then placed into master inoculum were required to target 6.0 log CFU/cm2 for each stomacher bags. The cooked roast beef was sampled as early as surface area. The calculated amount of E. coli O157:H7 was spot possible after thermal processing to prevent residual cooking. inoculated on the lean surface of the muscle. Top rounds were inoculated on three surfaces (two 8 by 8 cm and one 5 by 5 cm). Experiment 1: E. coli O157:H7 analysis. To analyze Top sirloins were inoculated on three surfaces (two 6 by 6 cm and surface inoculation counts, the samples (5 by 5 cm) were placed one 5 by 5 cm). Sterile aluminum templates were used as an into sterile stomacher bags with 225 ml of 0.1% peptone (Difco, inoculation guide, and the corners of the designated area were BD). The samples from each third of the raw and cooked beef were marked using dye (GL #80 Purple Brite, Great Lakes Meat weighed and placed in stomacher bags with 200 ml of E. coli Branding Ink, Koch Supplies Inc., Kansas City, MO). The medium (EC medium), modified (Difco, BD) containing novobi- inoculated beef was allowed to rest for 30 min, allowing the E. ocin antimicrobic supplement (Difco, BD). All samples were coli O157:H7 to attach. To calculate the initial E. coli O157:H7 stomached (Stomacher 400 circulator, Seward Laboratory Systems population on the inoculated beef surface, a portion (5 by 5 cm) of Inc., Bohemia, NY) for 2 min at 230 rpm. Serial dilutions were each muscle was aseptically excised using sterile scalpel handles made using 9 ml of 0.1% peptone. The brine samples were and blades prior to enhancement. analyzed using serial dilutions as above and were spiral plated (Autoplate 4000, Spiral Biotech, Norwood, MA) onto tryptic soy Experiment 1: subprimal injection and storage. Inoculated agar (BD, Franklin Lakes, NJ) containing 50 mg/ml of ampicillin subprimals were injected through the lean face with one of three (Sigma-Aldrich). Plates were incubated at 37 ¡ 2uC for 24 ¡ 2h solutions: 0.5% sodium chloride and 0.4% sodium tripolypho- and enumerated using a colony counter (Reichert Analytical sphate as a control (CNT); CNT with a 1% solution of 80% Instruments, Depew, NY). sodium citrate plus 20% sodium diacetate (SCzD) (IONAL LC, When E. coli O157:H7 was below the detection limit (1.6 | WTI Inc., Jefferson, GA), or CNT with 2% buffered vinegar (VIN) 101 CFU/g for 50 g of sample) the samples in EC medium were (MOstatin V, WTI Inc.) in the final product. A recirculating enriched and incubated at 37 ¡ 2uC for 24 ¡ 2 h. After 24 h, 10 ml multineedle injector with 21 needles (4-mm), operating at 41 of the EC medium from each sample was streaked onto tryptic soy strokes per min at 130 kPa (Injectamatic PI21, Koch Equipment agar plus ampicillin plates and incubated an additional 24 h at 37 LLC, Kansas City, MO) was calibrated using noninoculated ¡ 2uC. Suspected colonies were confirmed for E. coli O157 using muscles to deliver a 10% pickup. Brine solutions were made 12 h latex agglutination rapid detection kits (Dryspot E. coli O157 Latex prior to injection and mixed for 4 h. Immediately prior to use, the Kit, Oxoid Limited, Hampshire, UK). brine solution was remixed to ensure a homogenous solution. Prior to injection of treatment subprimals, two of the calibration Experiment 2: psychrotrophic bacteria. Similar procedures subprimals were surface inoculated (6.0 log CFU/cm2)to for meat procurement and subprimal injection were followed as in intentionally contaminate the brine and needles to ensure that all experiment 1, with the following changes. The subprimals were not treatment subprimals were exposed to similar conditions. Samples inoculated, and injections were conducted at the University of were collected from the brine solution: after mixing, after Georgia Meat Science and Technology Center (Athens). After contamination with calibration muscles, after injection of the fifth injection, subprimals were vacuum packaged (30 to 50 ml of O2/ subprimal, and after injection of the 10th subprimal. Brine pH was m2/24 h; 101,325 Pa; 23uC; B-620 series, Cryovac Sealed Air also measured before injection. Additionally, the pH and weight of Corporation) using a Type 800 vacuum packager (Original the subprimals were measured before and after injection. After Henkelman Vacuum Systems, Henkelman BV, The Netherlands) 2 injection, samples were vacuum packaged (30 to 50 ml of O2/m / and stored (0 ¡ 2uC) for 10 days. All equipment was thoroughly 24 h; 101,325 Pa; 23uC; B-620 series, Cryovac Sealed Air cleaned between each treatment to avoid mixing treatment Corporation, Duncan, SC) using a A300/16 (Multivac, Cryovac solutions. J. Food Prot., Vol. 74, No. 3 E. COLI AND PSYCHROTROPHIC BACTERIA INHIBITION IN BRINE-INJECTED BEEF 361

TABLE 1. Least square means of beef top round and top sirloin percent pickup, purge, and pHa Brine solutionb:

Characteristic CNT SCzD VIN

Top rounds Pickup, % 11.40 ¡ 0.55 10.28 ¡ 0.81 11.96 ¡ 0.62 Purge, % 1.76 ¡ 0.12 B 2.89 ¡ 0.27 A 2.84 ¡ 0.39 AB pH, before injection 5.71 ¡ 0.03 Z 5.72 ¡ 0.05 Z 5.70 ¡ 0.04 Z pH, after injection 6.11 ¡ 0.10 AY 5.79 ¡ 0.04 BYZ 5.92 ¡ 0.05 AB Y pH, after 10 days 5.71 ¡ 0.05 BZ 5.89 ¡ 0.05 AY 5.83 ¡ 0.05 AB Y Top sirloins Pickup, % 11.44 ¡ 0.55 10.26 ¡ 0.54 11.09 ¡ 0.64 ¡ B ¡ A ¡ A Purge, % 1.60 0.11 2.59 0.22 2.42 0.26 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 pH, before injection 5.62 ¡ 0.02 BZ 5.80 ¡ 0.03 AZ 5.82 ¡ 0.02 AZ pH, after injection 5.88 ¡ 0.06 Y 5.88 ¡ 0.04 Z 6.00 ¡ 0.05 Y pH, after 10 days 5.97 ¡ 0.04 AY 5.79 ¡ 0.04 BZ 5.74 ¡ 0.04 BZ a Values are means ¡ standard errors. Means within a row with different letters (A and B) differ (P , 0.05); means within a column, subprimal, and trait with different letters (Y and Z) differ (P , 0.05). b CNT, 0.5% sodium chloride and 0.4% sodium tripolyphosphate; SCzD, CNT with a 1% solution containing 80% sodium citrate plus 20% sodium diacetate; VIN, CNT with 2% buffered vinegar in the final product.

Experiment 2: steak fabrication and sampling. After BY statement. Differences among means were considered signi- 10 days of storage, each subprimal was fabricated into four steaks ficant at a # 0.05. ca. 2.54 cm thick. The steaks were randomly placed on absorbent pads (Dri-Loc AC-40, Cryovac Sealed Air Corporation) in RESULTS AND DISCUSSION polystyrene trays (Cryovac Sealed Air Corporation) and wrapped Experiment 1: E. coli O157:H7 challenge. There was with an oxygen-permeable polyvinylchloride overwrap (O2 transmission ~ 23,250 ml/m2/24 h, 72 gauge; Pro Pack Group, Oakland, no difference (P . 0.05) for percent brine pickup among z NJ). Steaks were stored at 4 ¡ 2uC with lighting to simulate a retail CNT, SC D, or VIN for top rounds or top sirloins display (960 lux) environment and were sampled for psychrotrophic (Table 1). The percent pickup averaged 11.21% for top organisms on days 1, 7, 14, and 21. To sample each steak, a section rounds and 10.93% for top sirloins. Percent purge after of surface area (5 by 5 cm) was aseptically excised to a depth of 10 days of vacuum storage exhibited a similar trend for top 0.2 cm. Samples were placed in sterile labeled stomacher bags. rounds and top sirloins. Subprimals that received CNT had 1.13 (P , 0.001) and 0.99% (P , 0.05) less purge than Experiment 2: psychrotrophic organism analysis. Two SCzD samples for top rounds and top sirloins, respective- hundred twenty-five milliliters of 0.1% peptone was added to each ly. However, purge loss was similar (P . 0.05) for CNT stomacher bag containing the sample (5 by 5 cm). The sample was compared with VIN and SCzD compared with VIN in top then stomached (Stomacher 400 circulator, Seward Laboratory rounds and top sirloins. Top round pH prior to injection was Systems Inc.) for 2 min at 230 rpm. Serial dilutions were made for similar (P . 0.05) among treatments. Immediately after all samples using 9 ml of 0.1% peptone. The dilutions were then injection, CNT pH was higher (P , 0.05) than SCzD, and spiral plated (Autoplate 4000, Spiral Biotech) onto plate count VIN was similar (P , 0.05) to both CNT and SCzD agar, and the plates were incubated at 4 ¡ 2uC for 7 days. Plates were counted using a colony counter (Reichert Analytical subprimals. After 10 days of storage, CNT pH was lower (P Instruments). , 0.05) than SCzD, and VIN was intermediate (P . 0.05) to both treatments. There was an immediate pH increase (P Statistical analysis. For experiment 1, the data for treatment , 0.01) for CNT- and VIN-injected samples. However, effects on percent pickup, purge, pH, and E. coli O157:H7 counts after 10 days of storage, CNT pH decreased (P , 0.01) to were analyzed as a completely randomized split-plot design with preinjected levels, whereas SCzD pH continued to increase the individual subprimals defined as the experimental unit and the (P , 0.05). Initial pH for top sirloins was lower (P , 0.05) observational unit. Subprimal was nested within replication, and for CNT than SCzD and VIN. However, after enhance- subprimal within treatment was considered the random variable. ment, CNT and VIN pH increased (P , 0.05), and all three The E. coli O157:H7 counts from each third of the uncooked and treatments were similar (P . 0.05). After 10 days of the cooked halves were converted to log CFU per gram. For vacuum storage, VIN pH decreased (P , 0.05) but was still experiment 2, psychrotrophic counts were analyzed as a similar (P . 0.05) to SCzD. Top sirloin pH from CNT completely randomized split-plot with subprimal as the whole plot and steak randomly assigned to day of display as the split- continued to numerically increase and was greater (P , z plot. The count from each sample was converted to log CFU per 0.05) than SC D- and VIN-enhanced samples after 10 days square centimeter. Data were analyzed using the SAS v.9.1 system of storage. In the current study, pH (initial, after (Proc Mixed; SAS Institute, Cary, NC). Means were separated enhancement, or after 10 days of storage) did not influence using the PDIFF option in LSMEANS with Tukey adjustment. (covariate analysis) E. coli O157:H7. Currently there is no When an interaction was present, data were reanalyzed with the literature describing how changes in meat pH after brine 362 PONRAJAN ET AL. J. Food Prot., Vol. 74, No. 3

TABLE 2. Least square means for surface E. coli O157:H7 inoculation level and E. coli O157:H7 recovered from the top, middle, and bottom thirds of enhanced raw beef top rounds and top sirloinsa Brine solutionb:

Location CNT SCzD VIN

Top rounds Inoculation level (log CFU/cm2) 6.40 ¡ 0.06 6.38 ¡ 0.05 6.40 ¡ 0.05 E. coli O157:H7 (log CFU/g) Top 1/3 5.46 ¡ 0.04 AY 5.06 ¡ 0.12 BY 5.15 ¡ 0.10 BY Middle 1/3 4.67 ¡ 0.17 AZ 3.74 ¡ 0.13 BZ 3.82 ¡ 0.11 BZ Bottom 1/3 4.47 ¡ 0.25 AZ 3.82 ¡ 0.09 BZ 3.75 ¡ 0.09 BZ Top sirloins

Inoculation level (log CFU/cm2) 6.72 ¡ 0.07 6.56 ¡ 0.04 6.56 ¡ 0.07 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 E. coli O157:H7 (log CFU/g) Top 1/3 5.98 ¡ 0.12 AY 5.68 ¡ 0.16 AB Y 5.39 ¡ 0.12 BY Middle 1/3 5.51 ¡ 0.15 AZ 4.07 ¡ 0.09 BZ 4.15 ¡ 0.14 BZ Bottom 1/3 5.31 ¡ 0.16 AZ 4.00 ¡ 0.12 BZ 4.07 ¡ 0.12 BZ a Values are means ¡ standard errors. Means within a row with different letters (A and B) differ (P , 0.05); means within a column, subprimal, and third with different letters (Y and Z) differ (P , 0.05). b CNT, 0.5% sodium chloride and 0.4% sodium tripolyphosphate; SCzD, CNT with a 1% solution containing 80% sodium citrate plus 20% sodium diacetate; VIN, CNT with 2% buffered vinegar in the final product. injection with antimicrobials influences the effectiveness of organisms increased (P , 0.05) until day 14 for CNT and the antimicrobials. until day 21 for SCzD and VIN samples (Table 3). Steaks During the injection of top rounds and top sirloins no E. enhanced with only sodium chloride and sodium tripolyphos- coli O157:H7 was detected in the brine solution prior to phate had higher levels of psychrotrophic organisms on days calibration. The E. coli O157:H7 level in the brine was 7, 14, and 21 compared with SCzD and VIN (P , 0.05). between 4.1 and 4.6 log CFU/ml for all treatments There was a mean difference in psychrotrophic populations of immediately after injecting the inoculated calibration 2.36 and 2.18 log CFU/cm2 between CNT and SCzDor subprimals, and it continued to increase between 4.7 and CNT and VIN steaks, respectively, across all days. Steaks 5.2 log CFU/ml after injection of the 5th and 10th from SCzD had fewer (P , 0.05) psychrotrophic organisms subprimals for both top rounds and top sirloins. than VIN on day 14 (P , 0.01), but there was no difference Top round E. coli O157:H7 counts from the top third of (P . 0.05) between SCzD and VIN on days 7 or 21. Similar CNT were greater (P , 0.05) than SCzD and VIN samples by to top rounds, extended aerobic storage of enhanced top 0.40 and 0.31 log CFU/g, respectively (Table 2). However, there sirloin steaks resulted in psychrotrophic bacteria increasing was no difference (P . 0.05) between SCzD and VIN. In the until day 14 for CNT and until day 21 for SCzD and VIN middle and bottom third of top rounds, CNT had greater mean E. samples. Top sirloin CNT steaks had a higher level of coli O157:H7 counts compared with SCzDandVIN(P , psychrotrophic organisms on all days of aerobic storage 0.05). There was a mean difference of 0.66 and 0.63 log CFU/g compared with SCzD and VIN (P , 0.05). There was a between the CNT versus SCzD and CNT versus VIN samples, mean difference in psychrotrophic populations of 1.88 and respectively. Injected top sirloin E. coli O157:H7 counts 2.33 log CFU/cm2 between CNT and SCzD or CNT and followed a similar trend as top round samples (Table 2). CNT VIN for all days. Psychrotrophic counts for VIN were lower top sirloin subprimals had more E. coli O157:H7 in the top third than SCzD steaks on day 7 (P , 0.01), but there was no when compared with VIN (P , 0.01), but there was no difference (P . 0.05) on the other sampling days. difference between CNT and SCzDorSCzDandVIN Beef destined for brine injection that is contaminated samples (P . 0.05). In the middle and bottom third, CNT with E. coli O157:H7 could pass contamination to the line, exhibited a greater mean E. coli O157:H7 count compared with needles, and brine, which could further contaminate SCzDorVIN(P , 0.05). There was a mean difference of 1.02 subsequent cuts. Once the brine solution becomes contam- and 1.06 log CFU/g between CNT versus SCzDandCNT inated, pathogens are not only translocated from the surface, versus VIN, respectively. Among the three treatments, the top but may be injected directly into the deep tissues. The third layer had greater E. coli O157:H7 counts compared with the inclusion of antimicrobial agents in brine solutions could be middle and bottom third layers (P , 0.05). E. coli O157:H7 was an effective method to decrease the risk of pathogen not detected in any of the thirds from the cooked beef top rounds transmission from cut to cut. Wicklund et al. (20) reported or top sirloins (data not shown; detection limit 1.20 log CFU/g). that, when using recirculated brines with inoculated strip steaks, steaks injected with fresh brine had greater counts Experiment 2: retail display challenge. Extended than steaks injected with recycled solutions. The study aerobic storage of top rounds showed that psychrotrophic revealed that each steak subjected to a particular injection J. Food Prot., Vol. 74, No. 3 E. COLI AND PSYCHROTROPHIC BACTERIA INHIBITION IN BRINE-INJECTED BEEF 363

TABLE 3. Least square means for psychrotrophic bacteria for reduction of E. coli K-12. The results from the current enhanced steaks from beef top rounds and top sirloins over 21 days study using a 1% solution of sodium citrate (80%) plus a of aerobic storage diacetate (20%) or buffered vinegar (2%) are similar to Brine solutionb: those from previous research. Storage The USDA-FSIS recommends cooking roast beef for z day CNT SC D VIN 12 min after the center temperature of the roast has reached Top rounds 60uC to achieve a 6.5 to 7.0 log reduction of Salmonella (14). The results from this research indicate that following 1 1.87 ¡ 0.15 Z 1.74 ¡ 0.09 Z 1.75 ¡ 0.08 Z 7 6.58 ¡ 0.10 AY 3.34 ¡ 0.14 BY 3.54 ¡ 0.13 BY USDA-FSIS guidelines for cook and hold times are also 14 9.47 ¡ 0.12 AX 5.76 ¡ 0.09 CX 6.20 ¡ 0.08 BX effective at eliminating E. coli O157:H7. In addition, Gill 21 9.63 ¡ 0.12 AX 7.27 ¡ 0.23 BW 7.33 ¡ 0.15 BW and McGinnis (5) cooked retail whole muscle, nonintact Top sirloins beef from 63 to 68uC and reported that cooking whole muscle, nonintact beef to medium rare would be adequate ¡ AZ ¡ BZ ¡ BZ

1 3.88 0.17 2.65 0.21 2.70 0.11 Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 7 8.20 ¡ 0.10 AY 5.46 ¡ 0.23 BY 4.42 ¡ 0.26 CY for reducing aerobic bacteria. Similarly, Luchansky et al. 14 9.46 ¡ 0.10 AX 7.66 ¡ 0.25 BX 7.02 ¡ 0.22 BX (10), using a commercial gas grill to cook E. coli O157:H7– 21 9.79 ¡ 0.05 AX 8.04 ¡ 0.18 BW 7.85 ¡ 0.09 BW inoculated whole muscle, nonintact beef steaks, reported a 2.6 to 4.2 log CFU/g reduction when the meat was cooked a Values are means ¡ standard errors presented in log CFU per between 48.9 and 60 C. The current study supports that square centimeter. Means within a row with different letters u cooking brine-injected roast beef to 60 C with a 12-min (A through C) differ (P , 0.05); means within a column and u subprimal with different letters (W through Z) differ (P , 0.05). hold under commercial conditions is an effective method for b CNT, 0.5% sodium chloride and 0.4% sodium tripolyphosphate; eliminating internalized E. coli O157:H7. SCzD, CNT with a 1% solution containing 80% sodium citrate Sallam (12) reported that dipping fresh salmon slices in plus 20% sodium diacetate; VIN, CNT with 2% buffered vinegar a 2.5% aqueous solution of sodium acetate, sodium lactate, in the final product. or sodium citrate could extend aerobic storage time by up to 42%, with sodium acetate having the greatest effect treatment had received a different level of E. coli followed by sodium lactate and sodium citrate. It was contamination. In the current study the needles and brine reported that the control samples had 1.92 to 3.26 log more were intentionally contaminated prior to the initiation of the psychrotrophic bacteria than the treated samples at the end study to ensure that each subprimal received the same of the shelf-life study. These results are comparable to the treatment. Furthermore, the inclusion of sodium citrate plus current study, where sodium citrate plus sodium diacetate sodium diacetate or buffered vinegar in the brine solution and buffered vinegar restricted the growth of psychrotrophic did not inhibit E. coli O157:H7 in the brine solution when bacteria over extended aerobic storage. Although aerobic compared with the control brine. storage of steaks for 21 days is not done, these data show Although the exact composition of MOstatin V is not that SCzD and VIN are effective at reducing the outgrowth known (18), the active ingredient of buffered vinegar is of psychrotrophic organisms in brine-injected top round and acetic acid. The effects of acetic acid and citric acid, the main ingredient of IONAL LC, have been shown to be top sirloin steaks. effective against E. coli O157:H7 (1). Typically, the main This study indicates that sodium citrate plus sodium method for inactivation of pathogens on brine-injected beef diacetate or buffered vinegar can be included in enhance- has been to apply an antimicrobial to the surface of beef cuts ment solutions for whole muscle, nonintact beef top rounds prior to injection. Surface decontamination has been shown and top sirloins as a hurdle against E. coli O157:H7. The use to be successful in reducing the E. coli O157:H7 load on of these antimicrobials in the brine solution will also control inoculated beef prior to brine injection by .1.0 log CFU/ the growth of psychrotrophic organisms during aerobic 100 cm2 (6). However, it was reported that 73 of 76 brine- storage of brine-enhanced steaks. Additionally, thermal injected samples still had internal positive results for E. coli processing of injected beef top round and top sirloin O157:H7 (6). Evaluating the effectiveness of surface subprimals for 12 min after reaching 60uC internally was decontamination on E. coli O157:H7–inoculated beef, effective at eliminating internalized E. coli O157:H7. Echeverry et al. (4) reported that 1.0 to 3.0 log CFU/g E. ACKNOWLEDGMENTS coli O157:H7 were transferred internally. Although significant reductions of E. coli O157:H7 were reported, this The Georgia Food Processing Advisory Council funded this research research identifies a potential need for antimicrobials to in cooperation with World Technology Ingredients Inc. and FPL Foods LLC. Li Ma and Larry Beuchat, Center for Food Safety, University of reach internal tissues. Furthermore, Paulson et al. (11) Georgia, Griffin, supported this project by providing the E. coli O157:H7 reported that including 3.0% sodium lactate plus 0.25% cultures. diacetate in a 0.3% salt and 0.3% phosphate brine solution provided a greater coliform reduction in brine-injected strip REFERENCES steaks than sodium lactate alone. Similarly, Wicklund et al. 1. Abdul-Raouf, U. M., L. R. Beuchat, and M. S. Ammar. 1993. (19) reported that sodium lactate or a combination of Survival and growth of Escherichia coli O157:H7 in ground, roasted sodium lactate and diacetate when incorporated in brines for beef as affected by pH, acidulants, and temperatures. Appl. Environ. inoculated strip steaks resulted in 1 to 2 log CFU/g Microbiol. 59:2364–2368. 364 PONRAJAN ET AL. J. Food Prot., Vol. 74, No. 3

2. Bell, B. P., M. Goldoft, P. M. Griffin, M. A. Davis, D. C. Gordon, P. 11. Paulson, D. D., R. A. Wicklund, M. C. Rojas, and M. S. Brewer. I. Tarr, C. A. Bartleson, J. H. Lewis, T. J. Barrett, J. G. Wells, R. 2007. Effects of shelf life enhancers and microbial load on Baron, and J. Kobayashi. 1994. A multistate outbreak of Escherichia Escherichia coli K12 survival in injected beef strip steaks. J. Muscle coli O157:H7-associated bloody diarrhea and hemolytic uremic Foods 18:194–206. syndrome from hamburgers—the Washington experience. JAMA (J. 12. Sallam, K. I. 2007. Antimicrobial and antioxidant effects of sodium Am. Med. Assoc.) 272:1349–1353. acetate, sodium lactate, and sodium citrate in refrigerated sliced 3. Doyle, M. P. 1991. Escherichia coli O157:H7 and its significance in salmon. Food Control 18:566–575. foods. Int. J. Food Microbiol. 12:289–302. 13. Sporing, S. B. 1999. Escherichia coli O157:H7 risk assessment for 4. Echeverry, A., J. C. Brooks, M. F. Miller, J. A. Collins, G. H. production and cooking of blade tenderized beef steaks. Master’s Loneragan, and M. M. Brashears. 2009. Validation of intervention thesis. Kansas State University, Manhattan. strategies to control Escherichia coli O157:H7 and Salmonella 14. U.S. Department of Agriculture, Food Safety and Inspection Service. Typhimurium DT 104 in mechanically tenderized and brine-enhanced 1999. Appendix A: Compliance guidelines for meeting lethality beef. J. Food Prot. 72:1616–1623. performance standards for certain meat and poultry products. 5. Gill, C. O., and J. C. McGinnis. 2004. Microbiological conditions of Available at: http://www.fsis.usda.gov/oa/fr/95033F-a.htm. Accessed mechanically tenderized beef cuts prepared at four retail stores. Int. J. 5 October 2009. Food Microbiol. 95:95–102. 15. U.S. Department of Agriculture, Food Safety and Inspection Service. 6. Heller, C. E., J. A. Scanga, J. N. Sofos, K. E. Belk, W. Warren-Serna, 1999. Beef products contaminated with Escherichia coli O157:H7. Downloaded from http://meridian.allenpress.com/jfp/article-pdf/74/3/359/1685103/0362-028x_jfp-10-294.pdf by guest on 01 October 2021 G. R. Bellinger, R. T. Bacon, M. L. Rossman, and G. C. Smith. 2007. Fed. Regist. 64:2803–2805. Decontamination of beef subprimal cuts intended for blade 16. U.S. Department of Agriculture, Food Safety and Inspection Service. tenderization or moisture enhancement. J. Food Prot. 70:1174–1180. 2004. Verification of procedures for controlling fecal material, 7. Jay, M. T., V. Garrett, J. C. Mohle-Boetani, M. Barros, J. A. Farrar, ingesta, and milk in slaughter operations. Directive 6420.2. USDA- R. Rios, S. Abbott, R. Sowadsky, K. Komatsu, R. Mandrell, J. Sobel, FSIS, Washington, DC. and S. B. Werner. 2004. A multistate outbreak of Escherichia coli 17. U.S. Department of Agriculture, Food Safety and Inspection Service. O157:H7 infection linked to consumption of beef tacos at a fast-food 2005. HACCP plan reassessment for mechanically tenderized beef restaurant chain. Clin. Infect. Dis. 39:1–7. products. Fed. Regist. 70:30331–30334. 8. Kennedy, J. E., S. K. Williams, T. Brown, and P. Minerich. 2006. 18. Valenzuela-Martinez, C., A. Pena-Ramos, V. K. Juneja, N. R. Prevalence of Escherichia coli O157:H7 and indicator organisms on Korasapati, D. E. Burson, and H. Thippareddi. 2010. Inhibition of the surface of intact subprimal beef cuts prior to further processing. J. Clostridium perfringens spore germination and outgrowth by Food Prot. 69:1514–1517. buffered vinegar and lemon juice concentrate during chilling of 9. Laine, E. S., J. M. Scheftel, D. J. Boxrud, K. J. Vought, R. N. Danila, ground turkey roast containing minimal ingredients. J. Food Prot. 73: K. M. Elfering, and K. E. Smith. 2005. Outbreak of Escherichia coli 470–476. O157:H7 infections associated with nonintact blade-tenderized frozen 19. Wicklund, R. A., D. D. Paulson, M. C. Rojas, and M. S. Brewer. steaks sold by door-to-door vendors. J. Food Prot. 68:1198– 2006. Effects of shelf-life enhancers on E. coli K12 survival in 1202. solutions used to enhance beef strip steaks. J. Food Sci. 71:M190– 10. Luchansky, J. B., A. C. S. Porto-Fett, B. Shoyer, R. K. Prebus, H. M195. Thippareddi, and J. E. Call. 2009. Thermal inactivation of 20. Wicklund, R. A., D. D. Paulson, M. C. Rojas, and M. S. Brewer. Escherichia coli O157:H7 in blade-tenderized beef steaks cooked 2007. The effects of shelf-life enhancers on E. coli K12 survival in on a commercial open-flame gas grill. J. Food Prot. 72:1404– needle-injected, surface contaminated beef strip steaks enhanced 1411. using recycled solutions. Meat Sci. 75:371–380.