Bactericidal Activity of Organic Acids against Salmonella typhimurium Attached to Broiler Chicken Skin t

Size: px

Start display at page:

Download "Bactericidal Activity of Organic Acids against Salmonella typhimurium Attached to Broiler Chicken Skin t"

Willa Daniels
5 years ago
Views:

1 629 Journal of Food Protection, Vol. 60, No.6, 1997, Pages Copyright, Intemational Association of Milk, Food and Environmental Sanitarians Bactericidal Activity of Organic s against Salmonella typhimurium Attached to Broiler Chicken Skin t KATHERINE C. TAMBLYN and DONALD E. CONNER* Auburn University, Poultry Science Department, Food Technology Institute, and Alabama Agricultural Experiment Station, Auburn, Alabama , USA (MS# : Received 20 August 1996/Accepted I November 1996) ABSTRACT The bactericidal activity of 0.5, 1,2,4, and 6% acetic, citric, lactic, malic, mandelic, propionic, or tartaric acid was determined against Salmonella typhimurium that were loosely or firmly attached to broiler chicken skin by using the skin-attachment model. treatments were applied during a simulated chill (0 C/60 min), postprocess dip (23 C for 15 s), or scald (50 C for 2 min). For comparison, activity of the acid treatments when applied under these conditions were also determined against S. typhimurium in aqueous suspension. In general, bactericidal activity (mean reduction log CPU per skin) of all acids increased linearly with increasing concentration in all applications. The bactericidal activity of organic acids depended on concentration and method of application. When compared to freely suspended cells, it is clear that salmonellae both firmly and loosely attached to poultry skin have increased resistance to or are protected from organic acids. In general, concentrations of ~4% of the acids were required to kill ~2 log number of cells of S. typhimurium that were attached to broiler skin. Key words: Salmonella, organic acids, bactericide, poultry, bacterial attachment Poultry have been identified as one of the most important reservoirs of salmonellae in the human food chain (4). Salmonellae can occur naturally in the intestinal tracts of chickens and can be widespread throughout both production and processing environments. Bacterial contamination of poultry carcasses during processing is undesirable and unavoidable. Chickens naturally can carry a variety of bacteria, including salmonellae, into the processing plant. The presence of salmonellae either on or in the live bird can result in transfer of these bacteria to the retail product. One United States Department of Agriculture Food Safety and Inspection Service study cited by Lillard (15) reported that 4% of the broilers entering the processing plant tested positive for salmonella, whereas * Author for correspondence: Tel: (334) ; Fax: (334) ; dconner@ag.auburn.edu t AAES Journal no % of the carcasses leaving the plant were salmonella positive. With up to 65% of retail carcasses testing positive for salmonella, emphasis must be placed at processing to reduce or eliminate salmonella contamination before the product reaches the consumer (1). A variety of antimicrobial treatments for broiler carcasses have been investigated, with primary focus on those that are practical and effective (6). Commonly used disinfectants include chlorine and short-chain organic acids. Both low (18 to 40 ppm) and high (>100 ppm) levels of chlorine may be effective in reducing bacterial counts on carcasses (12). The effectiveness of chlorine as a bactericidal agent is dependent upon the conditions in which it is used in the processing plant, including the concentration of chlorine, the temperature, and the chemical composition of the water (10, 12, 13, 16, 20). Low concentrations of chlorine may be effective in reducing bacterial counts on carcasses, but only if the volume of the chlorine solution in water per carcass is high and the relative amount of organic matter is low. This, however, would not be cost-effective in the commercial processing plant. Higher levels of chlorine are effective, but produce off flavor and carcass discoloration (19). They may also irritate the skin of plant personnel and are corrosive to plant equipment. Therefore, they are notuniversally acceptable for use throughout the poultry industry. Organic acids have been investigated because of their bactericidal activity and because they are generally recognized as safe (GRAS) and therefore are utilized for preservation in many food applications (5, 8, 20). In a study where 13 acids were evaluated, acetic and propionic were found to have the most inhibitory effect against salmonellae, whereas malic and lactic acids exhibited intermediate activity, and tartaric and citric were least inhibitory (2). Other research indicated that acetic acid was effective when used as a sanitizing spray or chill-water additive (6, 17). Further evaluation of organic acids may provide an economical and effective means of controlling salmonella contamination during processing. Presently, processing plants use carcass washes and immersion chill tanks, which are permitted to contain up to 50 ppm of a chlorine disinfectant (1). These methods often reduce but rarely eliminate salmonellae on poultry carcasses

2 630 TAMBLYN AND CONNER because they are ineffective against bacterial cells attached to the skin. Salmonellae that have become embedded in or firmly attached to the skin are more resistant to carcass treatments (14, 15, 19). This finding suggests a need for potential carcass treatments that will reduce or eliminate salmonellae firmly attached to broiler skin. The goal of this research was to evaluate the bactericidal activity of organic acids against Salmonella typhimurium attached to broiler skin. This research is unique among previous research conducted with organic acids because it utilized the skin-attachment model (SAM), an improved method for testing antimicrobial treatments against attached bacterial cells (3). The SAM utilizes irradiated chicken skin as a carrier for the test bacteria. Irradiation eliminates background microflora and permits testing of antimicrobial agents against a test bacterium in a pure culture situation. This allows recovery of bacteria in a nonselective medium, including surviving cells and those that may be sublethally injured. Also in this model, bacteria are allowed to attach to broiler skin prior to treatment with antimicrobial agents; therefore, a treatment's efficacy is determined against bacteria attached to or embedded in poultry skin. Specifically, the efficacy of 0.5, 1, 2, 4, or 6% acetic, citric, lactic, malic, mandelic, propionic, or tartaric acids against S. typhimurium attached to broiler skin during simulated chiller, postprocess dip, and scalder applications was determined. All organic acids, excluding mandelic acid, are GRAS and are commonly used food additives. MATERIALS AND METHODS Culture preparation Salmonella typhimurium, provided by N. A. Cox, USDA, Athens, GA. was used. Cultures were maintained at -80 C in brain heart infusion broth (BHI) (Difco Laboratories, Detroit, MI) containing 50% glycerol. Cultures were activated by two successive transfers in BHI at 23 C. Skin preparation Chicken breast skins were obtained from a local processor, cut into pieces (10 cm in diameter), and individually packaged in polyethylene Whirl-Pak bags (Nasco). Packaged skin samples were frozen (-20 C) and irradiated at ca. 10 kgy. Irradiated samples were stored at -20 C until used. Organic acid treatments Acetic (Fisher Scientific, Fair Lawn, NJ), Cltnc (Sigma Chemical Co., St. Louis, MO), lactic (Fisher), malic (Sigma), mandelic (Sigma), propionic (Sigma), and tartaric (Sigma) acids were prepared at 0.5, 1,2,4, and 6% (wt/wt). The ph of propionic acid was adjusted with 6 N HCI (Fisher) to fall within the ph range of all the other acids. The ph of the acid treatments ranged from 1.80 to 2.50, depending on the concentration and type of acid. Sterile water was used as a control for the experiment. Treatments were placed in either an ice water bath (O C), a hot water bath (50 C), or left at room temperature (23 C) prior to appropriate application (3). SAM experiments Irradiated chicken breast skin samples ("skins") were inoculated with 1 ml of S. typhimurium (10 5 CFU per skin) using a coarse aerosol spray (3). A IO-min contact time was allowed for bacterial cells to attach to the skin surface prior to application of treatments (3). Four skins (replicates) for each treatment were used. In accordance with commercial processing practices, treatments were applied at O C for 60 min to stimulate chilling, 23 C for 15 s to simulate a postprocess dip, and 50 C for 2 min to simulate scalding (4 replications per treatment per application). Following treatment, skins were placed in 20 ml of sterile 0.1 M phosphate buffer and shaken for 2 min. Samples of buffer were obtained to enumerate populations of S. typhimurium. Cells recovered at this point were identified as loosely attached cells. Skins were then transferred to 20 ml of fresh 0.1 M phosphate buffer and blended with a laboratory blender (Stomacher Model 400, Tekmar Co., Cincinnati, OH) for 2 min. Samples were obtained following blending to enumerate populations of S. typhimurium, which were defined as firmly attached cells. All samples were plated on tryptic soy agar (Difco) using a Spiral PlateJ'IW (Model DU2, Spiral Systems, Bethesda, MD), and were incubated at 37 C for 24 h. The original inoculum was enumerated using the same plating procedure. Expression of antibacterial activity Efficacy of the treatments was assessed by determining the reduction in viable populations of attached S. typhimurium. Reductions (log CFU per skin) were obtained by subtracting the number of cells recovered from the treated skin from the number of cells recovered from the inoculated skin treated with water (control). The difference was the reduction due to treatment. Statistical analysis Data, expressed as log CFU per skin, were subjected to an analysis of variance (7 by 5 factorial arrangement of organic acids and concentration) using SAS (18) software. Differences in means were separated using Tukey's studentized test (18). Data were analyzed by application and separately for loosely and firmly attached cell populations. Suspension tests For comparison, the activity of each treatment was determined against freely suspended S. typhimurium. S. typhimurium cells were added to each treatment for each application (100 ml) to provide an initial population of 10 7 CFU/ml. Treatments and applications were consistent with those used for the SAM experiments. Four replications per treatment per application were used. Samples were obtained following application and were plated using the plating method previously described. Reductions (log CFU/ml) were obtained by subtracting the population of cells recovered from each treatment from the population of cells recovered from the water control. RESULTS AND DISCUSSION Previously, it was determined in our laboratory that treatment effectiveness was affected by method of application (19). We observed that antimicrobial treatments were most effective in the scalder application, and least effective in the postprocess dip application. Therefore, data from the present study were analyzed separately for each application and presented accordingly. Except against loosely attached cells in the scalder application, a significant (P < 0.05) acid by concentration interaction was present for loosely and firmly attached cells (Table 1). The effects of the acid by concentration interaction on bactericidal activity (reduction log CPU per skin) against

3 BACTERICIDAL ACTIVITY OF ORGANIC ACIDS 631 TABLE 1. Probability values from analysis of variance of acid and concentration effects on recovery of Salmonella typhimurium cells loosely and firmly attached to broiler chicken skin samples P of effects of simulated applications on loose or firm cell attachment Chiller (O C,60 min) Dip (23 C, 15 s) Scalder (50 C, 2 min) Factor Loose Firm Loose Firm Loose Firm (A) Concentration (C) Interaction (A X C) Nsa Orthogonal contrasts for concentration Linear Quadratic NS NS NS NS NS Cubic NS NS NS NS NS Quartic NS.015 NS NS NS NS a Not significant (P > 0.05). loosely attached S. typhimurium cells in each application are summarized in Tables 2 to 4. In the chiller application (Table 2), the following ranges in bactericidal activity against loosely attached S. typhimurium at the tested concentrations were observed: acetic (0 to 0.58), citric (0 to 1.48), lactic (0.36 to 2.27), malic (0.44 to 2.74), mandelic (0.24 to 2.55), propionic (0.51 to 1.75), and tartaric (0 to 1.46). In the dip application (Table 3), the following ranges of bactericidal activity were observed: acetic (0 to 0.14), citric (0 to 0.13), lactic (0 to 0.73), malic (0 to 0.27), mandelic (0 to 0.78), propionic (0 to 0.59), and tartaric (0 for all concentrations). In the scalder application (Table 4), the following ranges of bactericidal activity were observed: acetic (0.67 to 1.79), citric (0.81 to 1.50), lactic (1.04 to 3.12), malic (0.71 to 1.46), mandelic (0.51 to 2.13), propionic (0.44 to 1.35), and tartaric (0.37 to 1.43). The effects of acid by concentration interaction of organic acids on bactericidal activity (reduction log CFU per skin) against firmly attached cells in each application are also summarized in Tables 2 to 4. In the chiller application (Table 2) the following ranges (reduction in population, log CFU per skin) were observed: acetic (0 to 0.70), citric (0 to 1.89), lactic (0.24 to 1.94), malic (0.78 to 2.21), mandelic (0.21 to 2.01), propionic (0.66 to 2.21), and tartaric (0.43 to 1.67). In the dip application (Table 3), the following ranges in activity were observed: acetic (0 to 0.27), citric (0 to 0.03), lactic (0 to 1.19), malic (0 for all concentrations), mandelic (0.31 to 1.84), propionic (0 to 1.65), and tartaric (0 to 0.20). In the scalder application (Table 4), the following ranges in activity were observed: acetic (0.91 to 2.41), citric (0.32 to 1.94), lactic (0.86 to 2.41), malic (0.79 to 1.80), mandelic (0.41 to 1.84), propionic (0.81 to 1.17), and tartaric (1.24 to 2.04). As stated above, an expected concentration effect was evident (Table 1). In general, a linear increase in bactericidal activity with increasing concentrations occurred (Table 1). A review of the data presented (Tables 2 to 4) indicates that at 0.5% and 1% some activity was observed, but concentrations ;::::2% were required to kill more than 100 salmonellae cells on a skin sample. Treatments were considered effective if they were able to reduce salmonellae levels by 2 log units, because salmonellae levels are typically known to be <100 CFU per carcass following processing (9). Against loosely attached S. typhimurium cells, 6% lactic acid and 6% mandelic when applied under chiller and scalder applications, and ;::::4% malic acid in the chiller application were TABLE 2. Bactericidal activity (reduction log CFU per skin sample) in the simulated chiller application (DOC,60 min) of various organic acids at increasing concentrations against Salmonella typhimurium cells loosely and firmly attached to broiler chicken skin samples Bactericidal effect (mean reduction log CPU S. typhimurium per skin sample) of organic acids on loosely and firmly attached cells (%) (0.075) (0.372) (0.224) (0.288) (0.228) (0.358) (0.214) (0.308) (0.221) (0.10) (0.235) (0.254) (0.184) (0.370) 0.5 OBb a 0.06c 0.30B 0.36c 0.67AB 0.44c 0.78B 0.24c 0.39B 0.51B 0.66B OB B 0.64 Oc OB 0.56c O.64AB 0.74BC 2.21A 0.38BC 0.56B 0.83AB 1.07B OB OB BC 0.44B 0.80BC 0.24B 1.45B 1.46AB 0.46BC 0.21B 1.13AB 1.53AB 0.26B A l0AB 1.85A 1.60AB 0.73AB 2.74A 2.21A 1.26B 2.01A 1.16AB 2.21A 0.46B A A 1.89A 2.27A 1.94A 2.74A 2.21A 2.55A 2.01A l.75a 2.21A 1.46A 1.67 a Values in parentheses are pooled standard error of the mean.

4 632 TAMBLYN AND CONNER TABLE 3. Bactericidal activity (reduction log CFU per skin sample) in the simulated postprocess dip application (23 C, 15 s) of various organic acids at increasing concentrations against Salmonella typhimurium cells loosely and firmly attached to broiler chicken skin samples Bactericidal effect (reduction log CPU S. typhimurium per skin sample) of organic acids on loosely and firmly attached cells (%) (O.lOl)a (0.124) (0.075) (0.070) (0.062) (0.350) (0.670) (0) (0.144) (0.338) (0.033) (0.237) (0) (0.072) Oc AB 0 Oc 0.03B OB 0.04B O.lOBc b AB BC 0.64AB OB OB Q BC 0 OB AB 0.31B 0.08B 0.21B B AB ABC 0.84A OB OB A A A 1.84A 0.59A 1.65A a Values in parentheses are pooled standard error of the mean. effective based on this criterion. Against firmly attached cells, 2::4% malic, mandelic, and propionic acids were effective in the chiller, whereas 2::4% lactic acid and 6% tartaric acid were effective in the scalder application. None of the acid treatments were effective in the dip application. These results are consistent with previous research that indicates high concentrations of organic acids, including those tested in the present study, are effective against bacterial contamination (5, 6). However, the higher concentrations of acids, which are usually most effective, produce undesirable carcass characteristics (11, 17). We observed undesirable carcass characteristics such as bleaching of the skin and off odor with our skin samples treated with 2::2% of the test acid. For comparison, the activity of each acid treatment was determined against freely suspended S. typhimurium. The following ranges of activity (mean reduction log CPU/ml) for all acid treatments at the following concentrations were observed: 0.5% (0.03 to 4.59), 1% (0.23 to 2::7.0), 2% (1.03 to 2::7.0), 4% (2.48 to 2::7.0), and 6% (4.75 to 2::7.0). In general, concentrations 2::2% resulted in reductions 2::7.0 log CPU/ml, and it was evident that the GRAS acids tested in the present study, even at 0.5%, have the ability to kill S. typhimurium if the bacteria are freely suspended, allowing the acids to contact them. Since a nonselective neutral ph medium and buffered diluent were used, it can be concluded the effects observed were bactericidal in nature. When compared to results utilizing the SAM, these data clearly indicate that salmonellae attached to poultry skin had increased resistance to carcass treatments or were protected against the bactericidal effect. The mode of action of organic acids in inhibiting microbial growth or killing cells may be related to changes in the acid-base equilibrium, proton donation, and interference in the production of energy by the cell (7). The inhibitory effect of the acid observed in this study was likely due, in part, to a high concentration of the undissociated molecule, which would be the prevalent form of the acid given the low ph values (1.8 to 2.5) of the treatments. Reduction or elimination of salmonellae on poultry carcasses before the retail product reaches the consumer should reduce the risk offoodbome salmonellosis. Eliminating pathogens from ready-to-cook poultry will continue to be a challenge to the poultry industry until effective and feasible treatments are developed. Organic acids, particularly at 2::4%, can effectively reduce salmonellae attached to broiler skin. However, improvements are needed to overcome cost and quality obstacles associated with applying acids at this high concentration. Further evaluation of TABLE 4. Bactericidal activity (reduction log CFU per skin sample) in the simulated scalder application (50 C, 2 min) of various organic acids at increasing concentrations against Salmonella typhimurium cells loosely and firmly attached to broiler chicken skin samples Bactericidal effect (reduction log CPU S. typhimurium of organic acids on loosely and firmly attached cells (%) (0.316)a (0.281) (0.204) (0.203) (0.228) (0.054) (0.314) (0.302) (0.344) (0.290) (0.306) (0.394) (0.370) (0.352) B B B B B B A A AB A A A AB A A A A B A A A a Values in parentheses are pooled standard error of the mean.

5 BACTERICIDAL ACTIVITY OF ORGANIC ACIDS 633 organic acids may provide an economical and effective means of controlling salmonellae during poultry processing. REFERENCES 1. Anonymous Reducing Salmonella during processing. Poult. Process. 4: Chung, K. C., and J. M. Goepfert Growth of Salmonella at low ph. J. Food Sci. 35: Conner, D. E., and S. F. Bilgili Skin attachment model for improved laboratory evaluation of potential carcass disinfectants for their efficacy against Salmonella attached to broiler skin. J. Food Prot. 57: D'Aoust, J. Y Salmonella, p In M. P. Doyle (ed.), Foodborne bacterial pathogens. Marcel Dekker, Inc., New York. 5. Dickson, J. S Acetic acid action on beef tissue surfaces contaminated with Salmonella typhimurium. J. Food Sci. 57: Dickson, J. S., and M. E. Anderson Microbiological decontamination of food animal carcasses by washing and sanitizing systems: a review. J. Food Prot. 55: Doores, S Organic acids, p In A. L. Branen and P. M. Davidson (ed.), Antimicrobials in foods. Marcel Dekker, Inc., New York. 8. Izat, A. L., M. Colberg, M. H. Adams, M. A. Reiber, and P. W. Waldroup Production and processing studies to reduce the incidence of salmonellae on commercial broilers. J. Food Prot. 52: Jetton, J. P., S. F. Bilgili, D. E. Conner, J. S. Kotrola, and M. A. Reiber Recovery of salmonellae from chilled broiler carcasses as affected by rinse media and enumeration method. J. Food Prot. 55: Kotula, A. w., G. J. Banwart, and J. A. Kinner Effect of postchill washing on bacterial counts of broiler chickens. Poult. Sci. 46: Kotula, K., and R. Thelappurate Microbiological and sensory attributes of retail cuts of beef treated with acetic and lactic acid solutions. J. Food Prot. 57: Lillard, H. S Effect on broiler carcasses and water of treating chiller water with chlorine or chlorine dioxide. Poult. Sci. 59: Lillard, H. S Improved chilling systems for poultry. Food Techno!. 36:58-{) Lillard, H. S Effect of surfactant or changes in ionic strength on the attachment of Salmonella typhimurium to poultry skin and muscle. J. Food Sci. 53: Lillard, H. S Factors affecting the persistence of Salmonella during the processing of poultry. J. Food Prot. 52: Mead, G. C., and N. L. Thomas Factors affecting the use of chlorine in the spin-chilling of eviscerated poultry. Br. Poult. Sci. 14: Mountney, G. J., and J. O'Malley s as poultry meat preservatives. Poult. Sci. 44: SAS Institute, Inc SAS user's guide, 6th ed. SAS Institute Inc., Cary,NC. 19. Tamblyn, K. C., D. E. Conner, S. F. Bilgili, and G. S. Hal! Utilization of the skin attachment model (SAM) to determine the antimicrobial activity of potential carcass treatments. Poult. Sci. 72(SI): Thomson, J. E., G. J. Banwart, D. H. Sanders, anda. J. Mercuri Effect of chlorine, antibiotics, 13-propiolactone, acids, and washing on Salmonella typhimurium on eviscerated fryer chickens. Poult. Sci. 46:

C.M. Harris, S.K. Williams 1. PhD Candidate Department of Animal Sciences Meat and Poultry Processing and Food Safety

C.M. Harris*, S.K. Williams* 1. PhD Candidate Department of Animal Sciences Meat and Poultry Processing and Food Safety The Antimicrobial Properties of a Vinegar-based Ingredient on Salmonella Typhimurium and Psychrotrophs inoculated in Ground Chicken Breast Meat and stored at 3±1 C for 7 days C.M. Harris*, S.K. Williams*