Temperature Function Integration and the Development and Metabolism of Poultry Spoilage Bacteria

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1978, p /78/36-65$2./ Copyright 1978 American Society for Microbiology Vol. 36, No. 5 Printed in U.S.A. Temperature Function Integration and the Development and Metabolism of Poultry Spoilage Bacteria H. B. DAUD,'* T. A. McMEEKIN,' AND JUNE OLLEY2 Department ofagricultural Science, University of Tasmania,' and Commonwealth Scientific and Industrial Research Organization, Division of Food Research, Tasmanian Food Research Unit,2 Hobart, Tasmania, 71, Australia Received for publication 24 August 1978 The rate of spoilage of chicken tissues, the development of spoilage bacteria, and the utilization of amino acids by spoilage bacteria as a function of temperature were more accurately described by the general spoilage curve of Olley and Ratkowsky (Food Technol. Aust. 25:66-73, 1973; Food Technol. N.Z. 8:13-17, 1973) than by the linear equation of Spencer and Baines (Food Technol. [Chicago] 18: , 1964). Remaining shelf life of poultry tissues may be predicted at temperatures up to 16 C by using a temperature function integrator which incorporates the general spoilage curve. Ingram and Dainty (2) reviewed a number of methods for the evaluation of spoilage and shelf life prediction of flesh foods. These included estimation of bacterial numbers, extract release volume, and various biochemical indexes. In addition, mathematical and graphical methods which relate the rate of spoilage to the temperature history of the product have been suggested by Spencer and Baines (8) and Olley and Ratkowsky (5, 6). Spencer and Baines (8) derived a formula relating spoilage rate to temperature given by: kt = ko (1 + Co), where kt is a rate of spoilage at temperature 6, ko is a rate of spoilage at C, and C is a constant for linear response. They suggested a linear response over the range of -1 to 25 C. This linear response was used by Ronsivalli and Charm (7) as the basis of a shelf life prediction slide rule which permits prediction of the remaining shelf life of cod held at storage temperatures of from 32 to 46 F (O to ca. 7.8C). A general spoilage curve for proteinaceous foods based on various data in the literature was developed by Olley and Ratkowsky (5, 6) to describe the relationship between temperature and relative spoilage rate. At temperatures above 8C the linear equation of Spencer and Baines (8) was not valid, and the Arrenhius equation more accurately described the spoilage rate-temperature relationship. An electronic integrator device to measure the temperature history of fish was developed by Nixon (Report on Fish Quality, Seminar no. 3, Fishing Industry Board, New Zealand, p , 1971), and its use has been described by Olley and Ratkowsky (5, 6) and Olley (4, 4a). This instrument was used in conjunction with a temperature gradient in- 65 cubator to examine the effect of temperature on the spoilage of chicken carcasses, poultry tissues (skin, leg, and breast muscle), and the metabolism of other substrates inoculated with poultry spoilage bacteria. MATERIALS AND METHODS Effect of temperature on the spoilage of chicken carcasses. The effect of storage temperature on the spoilage of chicken carcasses was examined at, 1, 5, 1, and 22C. Carcasses obtained immediately after processing were wrapped in permeable polyethylene film and stored at each temperature until spoilage was evident. Spoilage was assessed by detection of off odors, appearance of sliminess, and a bacterial population of 17 cells per cm2 of skin. Temperature gradient incubator studies. The effect of storage temperature on the spoilage flora of leg and breast tissue macerates of poultry, the rate of bacterial growth in nutrient broth, and the rate of utilization of individual amino acids were examined by using a temperature gradient incubator (Toyo Kagaku Sangyo Co., Ltd., Tokyo, Japan) over a temperature range of to 22 C. Spoilage of poultry tissues. The four types of poultry tissues studied were breast skin, leg skin, breast muscle, and leg muscle. Samples of skin (2 by 16 cm2) and muscle (2 g) were excised from breasts and legs of poultry carcasses immediately after processing. Each tissue type was homogenized in 2 ml of REP solution by using a Colworth Stomacher, and 1 ml of homogenate was added to the L-shaped incubator tubes previously equilibrated for 2 h. Daily inspection of the tissue macerates was made for evidence of spoilage. The criteria to determine spoilage were production of off odors and presence of a bacterial population of 17 cells per ml of suspension. Growth of spoilage bacteria in nutrient broth. One strain each of Pseudomonas putrefaciens, Pseudomonas group I, and Pseudomonas group II

2 VOL. 36, 1978 were grown in nutrient broth for 24 h at 22 C, and a loopful of each culture was inoculated into the previously equilibrated incubator tubes. Rate of growth was determined by the time required to reach a bacterial population of 17 cells per ml of broth. Estimates of numbers were made by spreading.1-nl samples on nutrient agar and incubating at 22C for 2 days. Utilization of amino acids by spoilage bacteria. Free amino acids and small-molecular-weight nitrogenous compounds are precursors of off odors in poultry tissues (1). Consequently, the rate of degradation of a representative of each type of amino acid was examined; L-argiuine (basic), L-asparagine (acid amine), L-aspartic (acid), L-proline (heterocyclic), L- serine (neutral), and L-tyrosine (aromatic) were prepared as.2 M solutions (except tyrosine, which was.5 M) in.1 M phosphate buffer (ph 7.) and filter sterilized through.45-,um filters. Samples (4.5 ml) of individual amino acid solutions were aseptically added to sterile incubator tubes and allowed to equilibrate for 2 h in the temperature gradient incubator. Bacterial strains isolated from spoiling chicken and known to rapidly metabolize particular amino acids were chosen (H. B. Daud, Honours thesis, University of Tasmania, Hobart, 1978). Washed cells were inoculated to a final concentration of =17 cells per ml. Control tubes were prepared containing cells with equivalent amounts of Ringer solution. The rate of amino acid utilization was assessed by thin-layer chromatography. Silica gel plates were spotted with 15-pl samples removed daily from the incubator in a progression from high to low temperatures. Plates were developed in butanol-acetic acid-water (6:15:25) solvent to a distance of 1 cm and visualized in ninhydrin (.2%, wt/vol, in acetone). Rate of utilization of amino acids was determined by disappearance of ninhydrin-positive material at each temperature. Computation of data. All determinations of the effect of temperature were converted to likely apparent activation energies (g) by using the integrated Arrhenius equation (4): logi (k2/k1) = (,u/2.33r) * (7T2 - Ti)/(TI * T2), where k, and k2 are rates of spoilage, growth, or amino acid utilization at absolute temperatures T1 and T2, respectively, R is the gas constant (1.987 calories [ca J] * degree-1/mol), and IL is the apparent activation energy. Intercepts and constants for a linear response to temperature based on the equation of Spencer and Baines were also calculated. Results for the various rates were converted into the relative rates where relative rate = rate at temperature t/rate at C, and the relative rates were compared with the curve of Olley and Ratkowsky (5, 6). RESULTS ANI) DISCUSSION The relative rates of microbial growth or metabolism at various temperatures compared with that at C were grouped to the nearest.5c, and the results are shown in Table 1. These indicated that there was no difference in relative rates between the growth and metabolism of bacteria on chicken tissues, in nutrient broth, or with individual amino acids. The results were POULTRY SPOILAGE BACTERIA 651 pooled, and the mean relative rates and standard errors are shown in Table 1. The values are plotted with the general spoilage curve of Olley and Ratkowsky (5, 6) in Fig. 1, which indicates the similar shape of the curves up to 16C. Above this temperature limited data obtained from 8 of the 14 systems studied indicated that the experimental curves continued to rise, deviating from the curve of Olley and Ratkowsky. However, data obtained in the temperature range of 2 to 25C for spoiling chicken muscle by J. T. Patterson (Queen's University, Belfast, Northern Ireland) cited by Olev et al. (Olev. Daud, and McMeekin, Temperature and seafood spoilage, Indo-Pacific Fisheries Commission Symposium on Fish Utilization, Technology and Marketing in the IPFC Region, Manila, Philippines, 8-11 March 1978) indicated the plateau effect predicted by the general spoilage curve at the relative rate of 9.. The time taken to reach the specified spoilage level at C for all systems studied is shown in Table 2, together with the predicted effect of temperature on spoilage with both the linear equation of Spencer and Baines (8) and the Arrhenius equation. The effect of temperature on the various systems studied was clearly not linear. The values obtained for the constant for linear response was outside the range of.25 to.36 obtained by Spencer and Baines (8), and the intercepts were often negative at C. The linear equation is, therefore, inappropriate for prediction of shelf life over the temperature range of to 22C. In contrast, the values obtained for apparent activation energies ranged from 15 to 25 kilocalories (kcal; ca. 63 to 15 kj) per mol, and the correlation coefficients for the reaction rates of the systems were highly significant when examined by the Arrhenius law (Table 2). The results for the 14 data sets were distributed around kcal (ca ± 3.4 kj) per mol. This value is close to one of the most common energy barriers (17 kcal [ca kj] per mol) in physiological processes and individual enzyme systems (3). Olley and Ratkowsky (6) calculated a value of 18 kcal (ca kj) per mol for the general spoilage curve over the range of 3 to 15C. The results, therefore, support the hypothesis of Olley and Ratkowsky and suggest that the Arrhenius equation is more suitable than the equation of Spencer and Baines (8) to describe the effect of storage temperature on the development and metabolism of microorganisms on poultry tissues and other substrates. The general spoilage curve of Olley and Ratkowsky (5, 6) has been incorporated into the circuitry of the temperature function integrator

3 652 DAUD, McMEEKIN, AND OLLEY APPL. ENVIRON. MICROBIOL. TABLE 1. Relative rate of spoilage (rate at temperature t/rate at C) of chicken substrates and growth and metabolism of individual bacterial species isolated from chickens Relative rate of: Relative rate Mean relative Temp Growth in nutrient by the equa- Chicken substrate spoilage (OC) broth Metabolism of single amino acids rate ± stan- tion of Olley L t dard ] error and Ratla I 12[3F tioK 1112[13t114 kowsky ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± L ± ± ± ± ± ± ± ± ± ± t a Code numbers are for systems as described in Table 2. L (Solid State Equipment Ltd., Lower Hutt, New Zealand). This reads out equivalent days of storage at C and allows a prediction to be made of the remaining shelf life at C. The present electronics simulate reasonably well the relative spoilage rates of poultry tissues at various temperatures up to 16 C, but further data sets are required to validate the values obtained in the range of 16 to 25 C. If these are substantiated, the electronics can be manipulated to simulate a poultry spoilage curve. In any case, storage in the higher temperature range would lead to rapid and organoleptically noticeable signs of deterioration. It should be noted that, although the relative rates of spoilage of various substrates are similar,

4 VOL. 36, 1978 POULTRY SPOILAGE BACTERIA , * TEMPERATURE *C FIG. 1. Comparison of mean relative rates ofgrowth and metabolism ofpoultry spoilage bacteria () with relative rates predicted by the general spoilage curve of Olley and Ratkowsky (). TABLE 2. -C a_ * O Values of Spencer and Baines constant and intercept and apparent activation energies for reaction rates versus temperature Time Values of the foltoie lowing by the equa- Values of the following by the Arrenreach tion of Spencer and hius equation: System speci- Baines: level at Apparent ac- Correla- Delevl Con- L tivation en- tion coef- grees of ys) stant C ergy ficient free- (days) (kcal/mol)a (r) b dom Spoilage of: 1. Chicken carcasses (78.5) Breast skin macerate (63.4) Leg skin macerate (65.1) Breast muscle macerate (7.9) Leg muscle macerate (73.9) Growth in nutrient broth of: 6. Alteromonasputrefaciens (84.4) Pseudomonas group I (83.5) Pseudomonas group II (89.) Utilization of: 9. Arginine by Pseudomonas group (65.5) II 1. Asparagine by Pseudomonas (11.6) group II 11. Aspartic acid Pseudomonas group (72.2) II 12. Proline by Pseudomonas group (74.7) III/IV (75.2) Serine by Pseudomonas group I (94.9) Tyrosine by an enteric type a Units for values in parentheses are kilojoules per mole. b Al r values were significant at the.1% level..

5 654 DAUD, McMEEKIN, AND OLLEY the actual time to reach a specified level will also depend on the nature of the substrates and the initial number of contaminants. The latter is illustrated by the time necessary for spoilage of chicken skin at C compared with muscle, which carried a much lower initial load of microorganisms. This information is, therefore, also required to allow accurate prediction of shelf life. Broiler carcasses produced under conditions of good manufacturing practice should have a shelf life of 7 days when stored at 5 C. From Fig. 1 the effect of storage at elevated temperatures can be determined; e.g., the rate of spoilage at 15 C will be approximately three times that at 5 C, and that at 1 C will be approximately twice that at 5 C. Thus it is evident that temperature is a major and reproducible factor determining the development and metabolism of poultry spoilage bacteria. It is suggested that temperature function integrators incorporating the curve of Olley and Ratkowsky or a modified poultry spoilage curve may be useful in monitoring the temperature history of nonfrozen poultry products and allow accurate prediction of remaining shelf life. ACKNOWLEDGMENTS The generous financial assistance of the Australian Chicken Meat Research Committee and the cooperation of Glenila Poultry Service, Hobart, Tasmania, are gratefully acknowledged. APPL. ENVIRON. MICROBIOIL. We thank M. A. Line, D. A. Ratkowsky, and C. J. Thomas for their assistance and constructive criticism of the work. LITERATURE CITED 1. Adamcic, M., D. S. Clark, and M. Yaguchi Effect of psychrotolerant bacteria on the amino acid content of chicken skin. J. Food Sci. 35: Ingram, M., and R. H. Dainty Changes caused by microbes in spoilage of meats. J. Appl. Bacteriol. 34: Morales, M. F A note on limiting reaction and temperature coefficients. J. Cell. Comp. Physiol. 3: Olley, J Temperature indicators, temperature integrators, temperature function integrators and the food spoilage chain, p In Towards an ideal refrigerated food chain, Annexe Bull. Int. Inst. Froid., Commissions C2, Dl, D2, D3, El, Melbourne, Australia. International Institute of Refrigeration, Paris. 4a.Olley, J Current status of the theory of the application of temperature indicators, temperature integrators and temperature function integrators to the food spoilage chain. Int. J. Refrig. 1: Olley, J., and D. A. Ratkowsky Temperature function integration and its importance in the storage and distribution of flesh foods above the freezing point. Food. Technol. Aust. 25: Olley, J., and D. A. Ratkowsky The role of temperature function integration in monitoring fish spoilage. Food. Technol. N.Z. 8: Ronsivalli, L. J., and S. E. Charm Spoilage and shelf life prediction of refrigerated fish. Mar. Fish. Rev. 37: Spencer, R., and C. R. Baines The effect of temperature on the spoilage of wet white fish. I. Storage at constant temperatures between -l1c and 25 C. Food. Technol. (Chicago) 18: