- Email: [email protected]
Effects of treatment with lysozyme and nisin on the microflora and sensory properties of commercial pork
International Journal of Food Microbiology 85 (2003) 259 – 267 www.elsevier.com/locate/ijfoodmicro
Effects of treatment with lysozyme and nisin on the microflora and sensory properties of commercial pork Frances M. Nattress *, Lynda P. Baker Meat Microbiology, Lacombe Research Centre, Agriculture and Agri-Food Canada, 6000 C & E Trail, Lacombe, Alberta, Canada T4L 1W1 Received 10 December 2001; received in revised form 1 October 2002; accepted 25 November 2002
Abstract A mixture of lysozyme and nisin at a ratio of 3:1 (w/w) and at a surface concentration of approximately 260 Ag/cm2 was effective in controlling the growth of lactic acid bacteria, lactic acid bacteria able to grow in the presence of acetate and Brochothrix thermosphacta on naturally contaminated pork loins that were stored in vacuum packages at 2 jC for up to 6 weeks. When loins were removed, cut into chops, and displayed in a retail display case, the efficacy of the antimicrobial mixture decreased with increasing display time. At most sampling times, bacterial numbers were lower in treated samples than in untreated samples. The numbers of Enterobacteriaceae were higher in treated samples than in untreated samples. The growth of Enterobacteriaceae may have been reduced as a result of antimicrobial activity of the lactic acid bacteria because when the growth of lactic acid bacteria was inhibited by the antimicrobial treatment, the Enterobacteriaceae were able to grow to higher numbers. Sensory evaluation of the loins showed no difference between treated and untreated samples, but aerobically displayed chops treated with antimicrobial had more prevalent off-odours and reduced odour acceptability than untreated samples. Crown Copyright D 2002 Published by Elsevier Science B.V. All rights reserved. Keywords: Lysozyme; Nisin; Pork
1. Introduction Access to the global marketplace and the meat industry’s requirement to extend storage life of fresh meat products has resulted in extensive use of vacuum packaging and refrigerated storage. Despite these hurdles, microbial spoilage of fresh meats still occurs and is a major source of financial losses to the
* Corresponding author. Tel.: +1-403-782-8140; fax: +1-403782-6120. E-mail address: [email protected] (F.M. Nattress).
industry. The application of nisin alone or in combination with other antimicrobials to meat surfaces has been shown to reduce numbers of some meat spoilage and/or pathogenic bacteria (Chung et al., 1989; Cutter and Siragusa, 1996a; Gill and Holley, 2000a; Holzapfel et al., 1995; Zhang and Mustapha, 1999). The incorporation of antimicrobials into edible coatings or packaging films and their efficacy at controlling growth of meat spoilage organisms and pathogens have also been studied (Cutter and Siragusa, 1996b, 1998; Gill and Holley, 2000b; Padgett et al., 1998). Nisin, produced by Lactococcus lactis subsp. lactis, a small, heat-stable protein, classified as a lanti-
0168-1605/02/$ - see front matter. Crown Copyright D 2002 Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-1605(02)00545-7
260
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
biotic (Holzapfel et al., 1995), is generally recognized as safe for use as a food additive in more than 50 countries (Delves-Broughton et al., 1996). It has been used extensively as an antimicrobial in different food products including processed cheeses and low- and high-acid canned foods. In areas of the world where ambient temperatures are high and refrigeration is poor, it has been used in pasteurized milk (DelvesBroughton et al., 1996). Its major limitations are that its spectrum of activity is limited to gram-positive bacteria and their spores (Delves-Broughton et al., 1996) and that it is very expensive. Nisin loses activity in fresh meat during storage (Cutter and Siragusa, 1994, 1996a,b, 1997; El-Khateib et al., 1993) probably due to an enzymatic reaction with glutathione as postulated by Rose et al. (1999). Lysozyme is a lytic enzyme found in foods such as milk and eggs. It is a muraminidase that hydrolyses h1,-4 linkages between N-acetylmuramic acid and Nacetylglucosamine. The action of lysozyme has been observed to be variable (Bester and Lombard, 1990) but gram-positive organisms tend to be more susceptible to lysozyme than gram-negative organisms, due to the lipoprotein –lipopolysaccharide layer in gramnegative organisms, rendering them less susceptible to disruption by lysozyme (Davidson et al., 1993). Its primary commercial application in foods has been in the control of Clostridium tyrobutyricum, the major cause of late blowing in semi-hard cheeses (Bester and Lombard, 1990; Cunningham et al., 1991). It has been suggested that the antimicrobial activity of lysozyme is attributed to both an enzymatic function and a catalytically independent activity (During et al., 1999; Ibrahim et al., 1996) and that denatured lysozyme may have increased activity against gram-negative bacteria (Ibrahim et al., 1996), particularly when accompanied by an additional stress such as high pressure (Masschalck et al., 2001). Recent studies have shown that combinations of lysozyme and nisin are synergistic (Chung and Hancock, 2000; Nattress et al., 2001). Chung and Hancock (2000) showed that a 3:1 combination of lysozyme and nisin was effective when minimal inhibitory concentrations for several meat spoilage organisms were determined. A study with inoculated pork confirmed that this ratio of lysozyme and nisin could inhibit the growth of a Carnobacterium sp. and B. thermosphacta and that it was more effective than
1:1 and 1:3 lysosyme-to-nisin ratios, lysozyme alone, and nisin alone (Nattress et al., 2001). This study was designed to evaluate the efficacy of a 3:1 combination of lysozyme and nisin as an antimicrobial treatment for commercial pork loins and chops cut from them.
2. Materials and methods 2.1. Antimicrobial The antimicrobial treatment used was a 3:1 lysozyme/nisin (w/w) mixture (Canadian Inovatech, Abbotsford, BC, Canada) at a concentration of 260 Ag/ cm2 of meat surface. The antimicrobial was dissolved in sterilized deionized water to a concentration of 10,000 Ag/ml of antimicrobial. It was estimated that an approximate concentration of antimicrobial upon the meat surface of 260 Ag/cm2 would be obtained by dipping the meat in this solution (Nattress et al., 2001). 2.2. Meat samples On three occasions, vacuum-packaged pork loins were obtained from a federally inspected abattoir on the day that they were processed. Within 1 h of being collected, the loins were cut in half. The pieces from each loin were assigned to treatments, either submerging in the antimicrobial solution or in sterile water for a period of 5 min. Eight loins per treatment were vacuum packaged (Multivac, Kansas City, MO, USA) in vacuum bags (Winpak, Winnipeg, MB, Canada) with an oxygen transmission rate of 40 – 50 cm3/m2/24 h/23 jC and were stored at 2 jC for up to 6 weeks. At 0 time and after 2, 4, and 6 weeks of storage, duplicate loins from each treatment were removed and opened. The surface of each loin was sampled in four locations using a sterile boring tool. The remaining portion of the loin was cut into chops. The chops were placed on Styrofoam trays (Scott National, Calgary, AB, Canada), which were overwrapped with an oxygen permeable, polyvinyl chloride film (Vitafilm Choice Wrap, Goodyear Canada, Toronto, ON, Canada) with an oxygen transmission rate of 8000 cm3/m2/24 h. Packaged chops were placed into a horizontal, retail display case (Model LPM12T, Hill Refrigeration of Canada, Barrie, ON, Canada) with fan-circulated air set to operate at a
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
261
blower temperature of 4 jC. The temperature on the surface of the meat was monitored using a Delphi temperature logger (Delphi Industries, Auckland, New Zealand). A detailed description and evaluation of this retail environment has been published (Greer and Jeremiah, 1980). Chops were displayed for up to 7 days. At 0 time and after 2, 5, and 7 days display in the retail case, two chops for each treatment were evaluated for their sensory and microbiological properties.
trained, five-member sensory panel (Greer et al., 1993). The odour case life was arbitrarily defined as the time at which the odour acceptability score rose to a value of 3, reflecting a moderate off-odour (Greer and Jones, 1996). Similarly, the retail case life was set as the time at which the retail appearance fell below 3.5.
2.3. Muscle pH, objective colour, and sensory evaluation
Four 10-cm2 cores were aseptically excised from each loin and two from each chop at each sampling period. Cores from each sample were combined, placed in 10 times their weight of 0.1% peptone diluent, and were homogenized for 2 min using a Colworth Stomacher (Baxter Diagnostics, Canlab Division, Edmonton, AB, Canada). Tenfold dilutions were prepared and samples were plated. The lower limit of sensitivity using this procedure was 1 log cfu/ cm2 for Enterobacteriaceae and 2.0 log cfu/cm2 for other bacteria. Selective media described by Baird et al. (1987) were used to enumerate pseudomonads, B. thermosphacta, lactic acid bacteria (LAB), and Enterobacteriaceae. Cephaloridine –fucidin – cetrimide agar (CFC) was used to enumerate pseudomonads. Plates were incubated for 2 days at 25 jC. The streptomycin sulfate – thallous acetate – actidione agar (STAA) plates used to enumerate B. thermosphacta were incubated for 2 days at 25 jC. Presumptive lactic acid bacteria and those LAB able to grow in the presence of 12 g/l acetate at pH 5.6 (AcLAB) were enumerated on MRS agar (Difco Laboratories, Detroit, MI, USA) and Rogosa agar (pH 5.6, Oxoid, Nepean, ON, Canada), respectively, with a 3-day anaerobic incubation at 25 jC. Anaerobic conditions were established using a BBL anaerobic system with an atmosphere containing 5 –10% CO2 (Becton Dick-
2.4. Microbiological analysis
The muscle pH was measured using an Oaktron Digital pH meter (Model Wo-0060500-000, Anachemia Scientific, Calgary, AB, Canada) equipped with a flat surface polymer body combination electrode (Fisher Scientific, Nepean, ON, Canada) and colour was objectively measured using a Minolta chromameter II (Minolta Camera, Japan). Triplicate readings of Commission Internationale de l’Eclairage (CIE, 1978) values (L*, a*, and b*) were recorded for each chop. Chroma (C*) and hue angle (h) were calculated using the equations C*=(a*2 + b*2)0.5 and h = arctan (b*/a*) where a* and b* are measured psychrometric chroma coordinates measuring the green to red axis and the blue to yellow axis, respectively. Sensory evaluation consisted of the assessment of muscle colour (five-point scale: 0 = completely discoloured; 1 = extremely pale; 5 = extremely dark), surface discolouration (seven-point scale: 1 = no surface discolouration; 7 = complete discolouration), retail appearance (seven-point scale: 1 = extremely undesirable; 7 = extremely desirable), off-odour intensity (five-point scale: 1 = no off-odour; 5 = prevalent offodour), and odour acceptability (five-point scale: 1 = acceptable; 5 = unacceptable) by an experienced,
Table 1 The pH and retail case life of pork chops, from loins that stored in vacuum for 2 and 6 weeks Time of loin storage (weeks)
pH Untreated
Treated
Appearance
Retail case life (days)
Untreated
Treated
Untreated
Treated
2 6
5.72 5.76
5.78 5.84
4 2
4 2
3.5 2
3 0.5
Odour
Loins had been treated by immersing them in a 3:1 aqueous solution of lysozyme and nisin, or were untreated.
262
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
generation of bubbles. Catalase-positive colonies were subcultured and Gram-stained using differential staining. Gram-positive and gram-negative cultures were included as controls. Subsequently, all MRS plates were flooded with 3% hydrogen peroxide and only catalase-negative colonies were counted as LAB. 2.6. Statistical analysis
Fig. 1. Number of colonies on MRS agar that were catalase negative (D) compared with the total number of colonies on MRS agar (x). Bacteria were isolated from samples of pork loins that had been dipped in sterile water (A) and from samples of pork loins that had been treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (B). Loins were stored vacuum packaged and stored at 2 jC for up to 6 weeks. The number of Enterobacteriaceae enumerated on Violet Red Bile Glucose Agar is also shown (dashed line).
inson, Cockeysville, MD, USA). Pour plates of Violet Red Bile Glucose Agar (Difco) with an overlay which were incubated for 18 to 24 h at 35 jC were used to enumerate Enterobacteriaceae. 2.5. Evaluation of the selectivity of MRS agar Colonies of different sizes from MRS were tested for catalase activity by transferring each colony to a drop of 3% hydrogen peroxide and observing the
Three replications were performed and data were analyzed using the General Linear Models Analysis of Variance Procedure of the Statistical Analysis System (SAS Institute, 1995). Means were compared by the Student’s t-test and differences were declared significant when P < 0.05. The model for loins included the main effects of treatment and week and their interaction. For chops the main effects, treatment, week and day, and all possible interactions were examined.
3. Results 3.1. Muscle pH, objective colour, and sensory evaluation The pH of the samples treated with antimicrobial was significantly higher ( P < 0.05) than untreated samples after both 2 and 6 weeks of storage (Table 1) but there was no difference immediately after antimicrobial treatment (data not shown).
Fig. 2. Number of LAB isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A, B, and C is 0.7. *Significant differences between treatments ( P < 0.05).
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
263
Fig. 3. Number of AcLAB that could grow in the presence of 12 g/l of acetate isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A is 0.3 and for B and C is 0.5. *Significant differences between treatments ( P < 0.05).
The antimicrobial treatment had no substantial effect on objective colour (CIE L*, a*, b*), subjective colour evaluation (colour, discolouration), or retail appearance. The retail case life for appearance of the product is shown in Table 1. The off-odour intensity and odour acceptability of the loins stored in vacuum were not affected by treatment and the product deteriorated over time, however, the antimicrobial treatment did have an impact on off-odour intensity and odour acceptability when chops were displayed in the retail case. The retail case life was reduced by up to 1.5 days on treated samples (Table 1). The average temperature
measured on the surface of the meat during retail display was 8.8 jC. 3.2. Bacterial numbers 3.2.1. Evaluation of the performance of MRS agar Fig. 1A shows differential counts from MRS agar based on the catalase reaction for untreated samples and Fig. 1B shows the counts for samples treated with antimicrobials. In treated samples, there was growth of a group of catalase-positive organisms, which were gram negative and, in the absence of a differential analysis, would mask the numbers of
Fig. 4. Numbers of B. thermosphacta isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A is 0.4 and for B and C is 0.5. *Significant differences between treatments ( P < 0.05).
264
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
Fig. 5. Number of Enterobacteriaceae isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A, B, and C is 0.3. *Significant differences between treatments ( P < 0.05).
LAB in the samples. In untreated samples, all colonies on MRS agar were catalase negative, typical of LAB. 3.2.2. Gram-positive bacteria The numbers of LAB in treated samples were significantly lower after 2, 4, and 6 weeks of storage ( P < 0.05) (Fig. 2A). After 6 weeks of storage, samples treated with nisin and lysozyme had 4.1 log units/cm2 fewer LAB than untreated samples. Similarly, enumeration of AcLAB demonstrated that their growth was almost eliminated when loins were treated with nisin and lysozyme (Fig. 3A). When chops were cut from the loins after different storage times, the differences between numbers of AcLAB on treated and untreated samples were maintained when chops were displayed for up to 7 days (Fig. 3B and C). LAB on treated samples were also lower but at some sampling times the differences were not significant (Fig. 2B and C). Growth of B. thermosphacta on treated samples was less than growth on untreated samples during storage in vacuum (Fig. 4A). When chops were cut from the loins and displayed in retail, at most sampling times the antimicrobial treatment had no effect on numbers of B. thermosphacta (Fig. 4B and C). 3.2.3. Gram-negative bacteria After 4 and 6 weeks of storage, the number of Enterobacteriaceae was significantly ( P < 0.05) higher in treated samples than in untreated samples (Fig. 5A).
When chops were cut from the loins, the growth rate of Enterobacteriaceae on both the untreated and treated samples was similar, but the number of these organisms on treated samples was higher at all sampling times and significantly higher at some sampling times (Fig. 5B and C). There was no effect of treatment on the pseudomonad numbers on stored loins or on chops cut from them (data not shown).
4. Discussion Although pH was higher during storage, immediately after the loins had been treated, their pH was only 0.1 pH unit higher than on untreated loins. It is likely, therefore, that the difference in pH was due to the different bacterial population that grew on treated samples during storage. The number of LAB and AcLAB was lower at all times on treated samples and products of their metabolism would be less than in untreated samples. When the off-odour intensity and odour acceptability were evaluated, the chops that were treated with antimicrobial became unacceptable earlier than untreated chops. Comments from the panelists suggest that the odour was putrid and quite different on the treated samples than the off-odours usually encountered. Among the bacterial groups enumerated, the only one that had higher numbers in treated samples were the Enterobacteriaceae and their numbers were
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
only about 0.5 log units higher in treated samples. Usually this group is not considered to be a major contributor to aerobic spoilage if glucose has not been exhausted (Gill, 1986). On loins, after 4 and 6 weeks of storage in vacuum, the number of Enterobacteriaceae was about 1 log unit higher on treated loins than on untreated loins. No perceptible difference in offodour intensity and odour acceptance was noted between the samples. It was only upon transfer of the samples to an aerobic environment and a higher temperature that differences in odour between treated and untreated samples were perceived. There is little information about the interaction of different organisms upon transfer from an anoxic environment to an aerobic one, particularly when the meat has been treated with an effective antimicrobial. When pork was transferred from an environment of 100% carbon dioxide to retail display, the LAB remained the dominant flora during retail display and growth of gram-negative bacteria was delayed (Greer et al., 1993). In this work, there was also a generation of off-odours which was not easily explained. A very detailed analysis of the bacterial population present when off-odours become apparent would be required to properly address this question. Early in these experiments, the data suggested that, except for an apparent increase in the population of Enterobacteriaceae, there was no effect of the antimicrobial mixture. When colonies on MRS agar were carefully observed, colonial differences were apparent, even though the total number of colonies on plates from treated and untreated samples was approximately the same. A further evaluation of the colonies demonstrated that, in the treated samples, the majority of the colonies on MRS agar were gram-negative and catalase positive, thus masking the efficacy of the antimicrobial against the LAB. Therefore, all colonies on MRS agar were differentiated using the catalase test and results were interpreted based on this reaction. When semi-selective media such as MRS or nonselective media are used in studies such as this one where the ‘‘natural’’ balance of the bacterial population is disrupted, it is possible that the effect of the antimicrobial treatment could be missed without careful observation of results. The enumeration of AcLAB whose numbers are usually 1 to 2 log units/cm2 lower than LAB enumerated on MRS confirms the efficacy of the antimicrobial and the difficulty with consider-
265
ing total counts on MRS agar to be those of presumptive LAB under these experimental conditions. The antimicrobial treatment evaluated controlled the growth of LAB and B. thermosphacta during 6 weeks of storage in vacuum. Treated samples had lower numbers of bacteria, which were not only statistically significant but also were greater than 1 log lower which would have practical significance as well. It is unusual for an antimicrobial treatment to inhibit these groups of bacteria for an extended storage period. When steaks were cut from loins, their initial numbers reflected the microbial condition of the loins. If the differences in numbers between untreated and treated samples were large, the differences on the steaks were greater than when the numbers on treated and untreated samples were more similar. During increased time in retail display, the numbers on treated samples generally increased to reach the same levels as untreated samples after 7 days. From a practical point of view, the effect of the antimicrobial combination was more pronounced during anoxic storage than during retail display. Most studies have been conducted on pure cultures and even with controlled conditions the efficacy of antimicrobials reduces significantly with increased storage time (Cutter and Siragusa, 1994, 1996a,b, 1997; El-Khateib et al., 1993). In previous work with the 3:1 mixture of lysozyme and nisin on sterile pork and pure bacterial cultures (Nattress et al., 2001), synergy between nisin and lysozyme was demonstrated. It was hypothesized that nisin had an immediate effect on the bacteria and that the antimicrobial effect was extended during storage by lysozyme or the mixture of lysozyme and nisin. Clearly, the 3:1 mixture of lysozyme and nisin was effective at controlling the growth of a variety of LAB that would be found on naturally contaminated pork. This includes extremely good activity against the AcLAB which have been associated with the development of off-flavours in beef stored in 100% carbon dioxide (Nattress and Jeremiah, 2000). If the spectrum of activity of these antimicrobials could be extended to include the gram-negative microflora and if they do not affect the flavour of the meat, they could be used in fresh meat systems. The spectrum of activity could be increased by altering the structure of the lysozyme (Ibrahim et al., 1994; Masschalck et al., 2001) or by covalently modifiying it (Ibrahim et al., 1994). Chelators such as EDTA or lactoferrin
266
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
could also be included in the antimicrobial mixture, although their activity in situ has not been promising (Cutter and Siragusa, 1995; Gill and Holley, 2000a). Alternatively, another antimicrobial such as lactate (Scannell et al., 1997), trisodium phosphate (Carneiro de Melo et al., 1998) or a polymerized amino acid (Shima et al., 1984; Ting et al., 1999) could be included in the mix that could potentially act synergistically with lysozyme and nisin. There has been considerable speculation and little data to describe what would happen to the microbial ecology when a group is inhibited. In this study, the inhibition of the LAB and the AcLAB appears to have resulted in a slight increase in the number of Enterobacteriaceae, and from the evaluation of odour attributes, possibly a shift in the population to strains that might cause early spoilage of the meat. Cudjoe (1988) reported a small increase in number of coliforms on cow heads that had been sprayed with lactic acid and stored in vacuum for 12 days when compared with unsprayed heads. Sensory evaluation did not differ between treated and untreated samples. Escherichia coli O157:H7 inoculated onto decontaminated and untreated beef grew to higher numbers on the decontaminated muscle than on the untreated muscle when samples were incubated for up to 15 days in vacuum at 10 jC, but the total number of bacteria was not different (Nissen et al., 2001). There is a need for a more detailed evaluation of the population shifts in the presence of this lysozyme/nisin antimicrobial mixture, particularly if an effort is undertaken to expand its spectrum of activity.
Acknowledgements Funding for this work was provided by Canadian Inovatech, Abbotsford, BC, Canada and the Matching Investment Initiative of Agriculture and Agri-Food Canada.
References Baird, R.M., Corry, J.E.L., Curtis, G.D.W. (Eds.), 1987. Pharmacopoeia of Culture Media for Food Microbiology. Int. J. Food Microbiol., vol. 5, pp. 187 – 299. Bester, B.H., Lombard, S.H., 1990. Influence of lysozyme on se-
lected bacteria associated with Gouda cheese. J. Food Prot. 53, 306 – 311. Carneiro de Melo, A.M., Cassar, C.A., Miles, R.J., 1998. Trisodium phosphate increases sensitivity of Gram-negative bacteria to lysozyme and nisin. J. Food Prot. 61, 839 – 843. Chung, W., Hancock, R.E., 2000. Action of lysozyme and nisin mixtures against lactic acid bacteria. Int. J. Food Microbiol. 60, 25 – 32. Chung, K.-T., Dickson, J.S., Crouse, J.D., 1989. Effects of nisin on growth of bacteria attached to meat. Appl. Environ. Microbiol. 55, 1329 – 1333. CIE (Commission Internationale de l’Eclairage), 1978. Recommendations on uniform color spaces – color difference equation – psychometric color terms. CIE publication no. 15 (E-1.3.1) 1891/(TC-1.3), supplement no. 2, Paris, France, pp. 8 – 12. Cudjoe, K.S., 1988. The effect of lactic acid sprays on the keeping qualities of meat during storage. Int. J. Food Microbiol. 7, 1 – 7. Cunningham, F.E., Proctor, V.A., Goetsch, S.J., 1991. Egg-white lysozyme as a food preservative: an overview. World’s Poult. Sci. J. 47, 141 – 163. Cutter, C.N., Siragusa, G.R., 1994. Decontamination of beef carcass tissue with nisin using a pilot scale model carcass washer. Food Microbiol. 11, 481 – 489. Cutter, C.N., Siragusa, G.R., 1995. Population reductions of Gramnegative pathogens following treatment with nisin and chelators under various conditions. J. Food Prot. 58, 977 – 983. Cutter, C.N., Siragusa, G.R., 1996a. Reductions of Listeria innocua and Brochothrix thermosphacta on beef following nisin spray treatments and vacuum packaging. Food Microbiol. 13, 23 – 33. Cutter, C.N., Siragusa, G.R., 1996b. Reduction of Brochothrix thermosphacta on beef surfaces following immobilization of nisin in calcium alginate gels. Lett. Appl. Microbiol. 23, 9 – 12. Cutter, C.N., Siragusa, G.R., 1997. Growth of Brochothrix thermosphacta in ground beef following treatments with nisin in calcium alginate gels. Food Microbiol. 14, 425 – 430. Cutter, C.N., Siragusa, G.R., 1998. Incorporation of nisin into a meat binding system to inhibit bacteria on beef surfaces. Lett. Appl. Microbiol. 27, 19 – 23. Davidson, P.M., Post, L.S., Branen, A.L., McCurdy, A.R., 1993. Naturally occurring and miscellaneous food antimicrobials. In: Branen, A.L., Davidson, P.M. (Eds.), Antimicrobials in Foods. Marcel Dekker, New York, pp. 371 – 419. Delves-Broughton, J., Blackburn, P., Evans, R.J., Hugenholtz, J., 1996. Applications of the bacteriocin nisin. Antonie van Leeuwenhoek 69, 193 – 202. During, K., Porsch, P., Mahn, A., Brinkmann, O., Gieffers, W., 1999. The non-enzymatic microbiocidal activity of lysozymes. FEBS Lett. 449, 93 – 100. El-Khateib, T., Yousef, A.E., Ockerman, H.W., 1993. Inactivation and attachment of Listeria monocytogenes on beef muscle treated with lactic acid and selected bacteriocins. J. Food Prot. 56, 29 – 33. Gill, C.O., 1986. The control of microbial spoilage in fresh meats. In: Pearson, A.M., Dutson, T.R. (Eds.), Advances in Meat Research, vol. 2. AVI Publishing, New York, pp. 49 – 88. Gill, A.O., Holley, R.A., 2000a. Inhibition of bacterial growth on
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267 ham and bologna by lysozyme, nisin and EDTA. Food Res. Int. 33, 83 – 91. Gill, A.O., Holley, R.A., 2000b. Surface application of lysozyme, nisin, and EDTA to inhibit spoilage and pathogenic bacteria on ham and bologna. J. Food Prot. 63, 1338 – 1346. Greer, G.G., Jeremiah, L.E., 1980. Influence of retail display temperature on psychrotrophic bacterial growth and beef case-life. J. Food Prot. 34, 542 – 546. Greer, G.G., Jones, S.D.M., 1996. Quality and bacteriological consequences of beef carcass spray-chilling: effects of spray duration and boxed beef storage temperature. Meat Sci. 45, 61 – 73. Greer, G.G., Dilts, B.D., Jeremiah, L.E., 1993. Bacteriology and retail case life of pork after storage in carbon dioxide. J. Food Prot. 56, 689 – 693. Holzapfel, W.H., Geisen, R., Schillinger, U., 1995. Biological preservation of foods with reference to protective cultures, bacteriocins and food-grade enzymes. Int. J. Food Microbiol. 24, 343 – 362. Ibrahim, H.R., Hatta, H., Fujiki, M., Kim, M., Yamamoto, T., 1994. Enhanced antimicrobial action of lysozyme against Gram-negative and Gram-positive bacteria due to modification with preillaldehyde. J. Agric. Food Chem. 42, 1813 – 1817. Ibrahim, H.R., Higashiguchi, S., Koketsu, M., Juneja, L.R., Kim, M., Yamamoto, T., Sugimoto, Y., Aoki, T., 1996. Partially unfolded lysozyme at neutral pH agglutinates and kills Gram-negative and Gram-positive bacteria through membrane damage mechanism. J. Agric. Food Chem. 44, 3799 – 3806. Masschalck, B., Van Houdt, R., Van Haver, E.G., Michiels, C.W., 2001. Inactivation of gram-negative bacteria by lysozyme, denatured lysozyme, and lysozyme-derived peptides under high hydrostatic pressure. Appl. Environ. Microbiol. 67, 339 – 344.
267
Nattress, F.M., Jeremiah, L.E., 2000. Bacterial mediated off-flavours in retail-ready beef after storage in controlled atmospheres. Food Res. Int. 33, 743 – 748. Nattress, F.M., Yost, C.K., Baker, L.P., 2001. Evaluation of the ability of lysozyme and nisin to control meat spoilage bacteria. Int. J. Food Microbiol. 70, 111 – 119. Nissen, H., Maugesten, T., Lea, P., 2001. Survival and growth of Escherichia coli O157:H7, Yersinia enterocolitica and Salmonella enteritidis on decontaminated and untreated meat. Meat Sci. 57, 291 – 298. Padgett, T., Han, I.Y., Dawson, P.L., 1998. Incorporation of foodgrade antimicrobial compounds into biodegradable packaging films. J. Food Prot. 61, 1330 – 1335. Rose, N.L., Sporns, P., Stiles, M.E., McMullen, L.M., 1999. Inactivation of nisin by glutathione in fresh meat. J. Food Sci. 64, 759 – 762. SAS Institute, 1995. SAS/STAT User’s Guide. SAS Institute, Cary, NC. Scannell, A.G.M., Hill, C., Buckely, D.J., Arendt, E.K., 1997. Determination of the influence or organic acids and nisin on shelflife and microbiological safety aspects of fresh pork sausage. J. Appl. Microbiol. 83, 407 – 412. Shima, S., Matsuoka, H., Iwamoto, T., Sakai, H., 1984. Antimicrobial action of a-poly-L-lysine. J. Antibiot. 37, 1449 – 1455. Ting, H.Y., Ishizaki, S., Tanaka, M., 1999. a-Polylysine improves the quality of surimi products. J. Muscle Foods 10, 279 – 294. Zhang, S., Mustapha, A., 1999. Reduction of Listeria monocytogenes and Escherichia coli O157:H7 numbers on vacuum-packaged fresh beef treated with nisin or nisin combined with EDTA. J. Food Prot. 62, 1123 – 1127.
Effects of treatment with lysozyme and nisin on the microflora and sensory properties of commercial pork Frances M. Nattress *, Lynda P. Baker Meat Microbiology, Lacombe Research Centre, Agriculture and Agri-Food Canada, 6000 C & E Trail, Lacombe, Alberta, Canada T4L 1W1 Received 10 December 2001; received in revised form 1 October 2002; accepted 25 November 2002
Abstract A mixture of lysozyme and nisin at a ratio of 3:1 (w/w) and at a surface concentration of approximately 260 Ag/cm2 was effective in controlling the growth of lactic acid bacteria, lactic acid bacteria able to grow in the presence of acetate and Brochothrix thermosphacta on naturally contaminated pork loins that were stored in vacuum packages at 2 jC for up to 6 weeks. When loins were removed, cut into chops, and displayed in a retail display case, the efficacy of the antimicrobial mixture decreased with increasing display time. At most sampling times, bacterial numbers were lower in treated samples than in untreated samples. The numbers of Enterobacteriaceae were higher in treated samples than in untreated samples. The growth of Enterobacteriaceae may have been reduced as a result of antimicrobial activity of the lactic acid bacteria because when the growth of lactic acid bacteria was inhibited by the antimicrobial treatment, the Enterobacteriaceae were able to grow to higher numbers. Sensory evaluation of the loins showed no difference between treated and untreated samples, but aerobically displayed chops treated with antimicrobial had more prevalent off-odours and reduced odour acceptability than untreated samples. Crown Copyright D 2002 Published by Elsevier Science B.V. All rights reserved. Keywords: Lysozyme; Nisin; Pork
1. Introduction Access to the global marketplace and the meat industry’s requirement to extend storage life of fresh meat products has resulted in extensive use of vacuum packaging and refrigerated storage. Despite these hurdles, microbial spoilage of fresh meats still occurs and is a major source of financial losses to the
* Corresponding author. Tel.: +1-403-782-8140; fax: +1-403782-6120. E-mail address: [email protected] (F.M. Nattress).
industry. The application of nisin alone or in combination with other antimicrobials to meat surfaces has been shown to reduce numbers of some meat spoilage and/or pathogenic bacteria (Chung et al., 1989; Cutter and Siragusa, 1996a; Gill and Holley, 2000a; Holzapfel et al., 1995; Zhang and Mustapha, 1999). The incorporation of antimicrobials into edible coatings or packaging films and their efficacy at controlling growth of meat spoilage organisms and pathogens have also been studied (Cutter and Siragusa, 1996b, 1998; Gill and Holley, 2000b; Padgett et al., 1998). Nisin, produced by Lactococcus lactis subsp. lactis, a small, heat-stable protein, classified as a lanti-
0168-1605/02/$ - see front matter. Crown Copyright D 2002 Published by Elsevier Science B.V. All rights reserved. doi:10.1016/S0168-1605(02)00545-7
260
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
biotic (Holzapfel et al., 1995), is generally recognized as safe for use as a food additive in more than 50 countries (Delves-Broughton et al., 1996). It has been used extensively as an antimicrobial in different food products including processed cheeses and low- and high-acid canned foods. In areas of the world where ambient temperatures are high and refrigeration is poor, it has been used in pasteurized milk (DelvesBroughton et al., 1996). Its major limitations are that its spectrum of activity is limited to gram-positive bacteria and their spores (Delves-Broughton et al., 1996) and that it is very expensive. Nisin loses activity in fresh meat during storage (Cutter and Siragusa, 1994, 1996a,b, 1997; El-Khateib et al., 1993) probably due to an enzymatic reaction with glutathione as postulated by Rose et al. (1999). Lysozyme is a lytic enzyme found in foods such as milk and eggs. It is a muraminidase that hydrolyses h1,-4 linkages between N-acetylmuramic acid and Nacetylglucosamine. The action of lysozyme has been observed to be variable (Bester and Lombard, 1990) but gram-positive organisms tend to be more susceptible to lysozyme than gram-negative organisms, due to the lipoprotein –lipopolysaccharide layer in gramnegative organisms, rendering them less susceptible to disruption by lysozyme (Davidson et al., 1993). Its primary commercial application in foods has been in the control of Clostridium tyrobutyricum, the major cause of late blowing in semi-hard cheeses (Bester and Lombard, 1990; Cunningham et al., 1991). It has been suggested that the antimicrobial activity of lysozyme is attributed to both an enzymatic function and a catalytically independent activity (During et al., 1999; Ibrahim et al., 1996) and that denatured lysozyme may have increased activity against gram-negative bacteria (Ibrahim et al., 1996), particularly when accompanied by an additional stress such as high pressure (Masschalck et al., 2001). Recent studies have shown that combinations of lysozyme and nisin are synergistic (Chung and Hancock, 2000; Nattress et al., 2001). Chung and Hancock (2000) showed that a 3:1 combination of lysozyme and nisin was effective when minimal inhibitory concentrations for several meat spoilage organisms were determined. A study with inoculated pork confirmed that this ratio of lysozyme and nisin could inhibit the growth of a Carnobacterium sp. and B. thermosphacta and that it was more effective than
1:1 and 1:3 lysosyme-to-nisin ratios, lysozyme alone, and nisin alone (Nattress et al., 2001). This study was designed to evaluate the efficacy of a 3:1 combination of lysozyme and nisin as an antimicrobial treatment for commercial pork loins and chops cut from them.
2. Materials and methods 2.1. Antimicrobial The antimicrobial treatment used was a 3:1 lysozyme/nisin (w/w) mixture (Canadian Inovatech, Abbotsford, BC, Canada) at a concentration of 260 Ag/ cm2 of meat surface. The antimicrobial was dissolved in sterilized deionized water to a concentration of 10,000 Ag/ml of antimicrobial. It was estimated that an approximate concentration of antimicrobial upon the meat surface of 260 Ag/cm2 would be obtained by dipping the meat in this solution (Nattress et al., 2001). 2.2. Meat samples On three occasions, vacuum-packaged pork loins were obtained from a federally inspected abattoir on the day that they were processed. Within 1 h of being collected, the loins were cut in half. The pieces from each loin were assigned to treatments, either submerging in the antimicrobial solution or in sterile water for a period of 5 min. Eight loins per treatment were vacuum packaged (Multivac, Kansas City, MO, USA) in vacuum bags (Winpak, Winnipeg, MB, Canada) with an oxygen transmission rate of 40 – 50 cm3/m2/24 h/23 jC and were stored at 2 jC for up to 6 weeks. At 0 time and after 2, 4, and 6 weeks of storage, duplicate loins from each treatment were removed and opened. The surface of each loin was sampled in four locations using a sterile boring tool. The remaining portion of the loin was cut into chops. The chops were placed on Styrofoam trays (Scott National, Calgary, AB, Canada), which were overwrapped with an oxygen permeable, polyvinyl chloride film (Vitafilm Choice Wrap, Goodyear Canada, Toronto, ON, Canada) with an oxygen transmission rate of 8000 cm3/m2/24 h. Packaged chops were placed into a horizontal, retail display case (Model LPM12T, Hill Refrigeration of Canada, Barrie, ON, Canada) with fan-circulated air set to operate at a
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
261
blower temperature of 4 jC. The temperature on the surface of the meat was monitored using a Delphi temperature logger (Delphi Industries, Auckland, New Zealand). A detailed description and evaluation of this retail environment has been published (Greer and Jeremiah, 1980). Chops were displayed for up to 7 days. At 0 time and after 2, 5, and 7 days display in the retail case, two chops for each treatment were evaluated for their sensory and microbiological properties.
trained, five-member sensory panel (Greer et al., 1993). The odour case life was arbitrarily defined as the time at which the odour acceptability score rose to a value of 3, reflecting a moderate off-odour (Greer and Jones, 1996). Similarly, the retail case life was set as the time at which the retail appearance fell below 3.5.
2.3. Muscle pH, objective colour, and sensory evaluation
Four 10-cm2 cores were aseptically excised from each loin and two from each chop at each sampling period. Cores from each sample were combined, placed in 10 times their weight of 0.1% peptone diluent, and were homogenized for 2 min using a Colworth Stomacher (Baxter Diagnostics, Canlab Division, Edmonton, AB, Canada). Tenfold dilutions were prepared and samples were plated. The lower limit of sensitivity using this procedure was 1 log cfu/ cm2 for Enterobacteriaceae and 2.0 log cfu/cm2 for other bacteria. Selective media described by Baird et al. (1987) were used to enumerate pseudomonads, B. thermosphacta, lactic acid bacteria (LAB), and Enterobacteriaceae. Cephaloridine –fucidin – cetrimide agar (CFC) was used to enumerate pseudomonads. Plates were incubated for 2 days at 25 jC. The streptomycin sulfate – thallous acetate – actidione agar (STAA) plates used to enumerate B. thermosphacta were incubated for 2 days at 25 jC. Presumptive lactic acid bacteria and those LAB able to grow in the presence of 12 g/l acetate at pH 5.6 (AcLAB) were enumerated on MRS agar (Difco Laboratories, Detroit, MI, USA) and Rogosa agar (pH 5.6, Oxoid, Nepean, ON, Canada), respectively, with a 3-day anaerobic incubation at 25 jC. Anaerobic conditions were established using a BBL anaerobic system with an atmosphere containing 5 –10% CO2 (Becton Dick-
2.4. Microbiological analysis
The muscle pH was measured using an Oaktron Digital pH meter (Model Wo-0060500-000, Anachemia Scientific, Calgary, AB, Canada) equipped with a flat surface polymer body combination electrode (Fisher Scientific, Nepean, ON, Canada) and colour was objectively measured using a Minolta chromameter II (Minolta Camera, Japan). Triplicate readings of Commission Internationale de l’Eclairage (CIE, 1978) values (L*, a*, and b*) were recorded for each chop. Chroma (C*) and hue angle (h) were calculated using the equations C*=(a*2 + b*2)0.5 and h = arctan (b*/a*) where a* and b* are measured psychrometric chroma coordinates measuring the green to red axis and the blue to yellow axis, respectively. Sensory evaluation consisted of the assessment of muscle colour (five-point scale: 0 = completely discoloured; 1 = extremely pale; 5 = extremely dark), surface discolouration (seven-point scale: 1 = no surface discolouration; 7 = complete discolouration), retail appearance (seven-point scale: 1 = extremely undesirable; 7 = extremely desirable), off-odour intensity (five-point scale: 1 = no off-odour; 5 = prevalent offodour), and odour acceptability (five-point scale: 1 = acceptable; 5 = unacceptable) by an experienced,
Table 1 The pH and retail case life of pork chops, from loins that stored in vacuum for 2 and 6 weeks Time of loin storage (weeks)
pH Untreated
Treated
Appearance
Retail case life (days)
Untreated
Treated
Untreated
Treated
2 6
5.72 5.76
5.78 5.84
4 2
4 2
3.5 2
3 0.5
Odour
Loins had been treated by immersing them in a 3:1 aqueous solution of lysozyme and nisin, or were untreated.
262
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
generation of bubbles. Catalase-positive colonies were subcultured and Gram-stained using differential staining. Gram-positive and gram-negative cultures were included as controls. Subsequently, all MRS plates were flooded with 3% hydrogen peroxide and only catalase-negative colonies were counted as LAB. 2.6. Statistical analysis
Fig. 1. Number of colonies on MRS agar that were catalase negative (D) compared with the total number of colonies on MRS agar (x). Bacteria were isolated from samples of pork loins that had been dipped in sterile water (A) and from samples of pork loins that had been treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (B). Loins were stored vacuum packaged and stored at 2 jC for up to 6 weeks. The number of Enterobacteriaceae enumerated on Violet Red Bile Glucose Agar is also shown (dashed line).
inson, Cockeysville, MD, USA). Pour plates of Violet Red Bile Glucose Agar (Difco) with an overlay which were incubated for 18 to 24 h at 35 jC were used to enumerate Enterobacteriaceae. 2.5. Evaluation of the selectivity of MRS agar Colonies of different sizes from MRS were tested for catalase activity by transferring each colony to a drop of 3% hydrogen peroxide and observing the
Three replications were performed and data were analyzed using the General Linear Models Analysis of Variance Procedure of the Statistical Analysis System (SAS Institute, 1995). Means were compared by the Student’s t-test and differences were declared significant when P < 0.05. The model for loins included the main effects of treatment and week and their interaction. For chops the main effects, treatment, week and day, and all possible interactions were examined.
3. Results 3.1. Muscle pH, objective colour, and sensory evaluation The pH of the samples treated with antimicrobial was significantly higher ( P < 0.05) than untreated samples after both 2 and 6 weeks of storage (Table 1) but there was no difference immediately after antimicrobial treatment (data not shown).
Fig. 2. Number of LAB isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A, B, and C is 0.7. *Significant differences between treatments ( P < 0.05).
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
263
Fig. 3. Number of AcLAB that could grow in the presence of 12 g/l of acetate isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A is 0.3 and for B and C is 0.5. *Significant differences between treatments ( P < 0.05).
The antimicrobial treatment had no substantial effect on objective colour (CIE L*, a*, b*), subjective colour evaluation (colour, discolouration), or retail appearance. The retail case life for appearance of the product is shown in Table 1. The off-odour intensity and odour acceptability of the loins stored in vacuum were not affected by treatment and the product deteriorated over time, however, the antimicrobial treatment did have an impact on off-odour intensity and odour acceptability when chops were displayed in the retail case. The retail case life was reduced by up to 1.5 days on treated samples (Table 1). The average temperature
measured on the surface of the meat during retail display was 8.8 jC. 3.2. Bacterial numbers 3.2.1. Evaluation of the performance of MRS agar Fig. 1A shows differential counts from MRS agar based on the catalase reaction for untreated samples and Fig. 1B shows the counts for samples treated with antimicrobials. In treated samples, there was growth of a group of catalase-positive organisms, which were gram negative and, in the absence of a differential analysis, would mask the numbers of
Fig. 4. Numbers of B. thermosphacta isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A is 0.4 and for B and C is 0.5. *Significant differences between treatments ( P < 0.05).
264
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
Fig. 5. Number of Enterobacteriaceae isolated from pork loins (A) that were vacuum packaged and stored at 2 jC for up to 6 weeks and for pork chops that were cut from the loins after 2 (B) and 6 (C) weeks of storage and displayed in a retail case set to operate at 4 jC. The loins were treated with lysozyme and nisin in a 3:1 ratio dissolved in sterile water (D) or were dipped in sterile water (x). The standard error for A, B, and C is 0.3. *Significant differences between treatments ( P < 0.05).
LAB in the samples. In untreated samples, all colonies on MRS agar were catalase negative, typical of LAB. 3.2.2. Gram-positive bacteria The numbers of LAB in treated samples were significantly lower after 2, 4, and 6 weeks of storage ( P < 0.05) (Fig. 2A). After 6 weeks of storage, samples treated with nisin and lysozyme had 4.1 log units/cm2 fewer LAB than untreated samples. Similarly, enumeration of AcLAB demonstrated that their growth was almost eliminated when loins were treated with nisin and lysozyme (Fig. 3A). When chops were cut from the loins after different storage times, the differences between numbers of AcLAB on treated and untreated samples were maintained when chops were displayed for up to 7 days (Fig. 3B and C). LAB on treated samples were also lower but at some sampling times the differences were not significant (Fig. 2B and C). Growth of B. thermosphacta on treated samples was less than growth on untreated samples during storage in vacuum (Fig. 4A). When chops were cut from the loins and displayed in retail, at most sampling times the antimicrobial treatment had no effect on numbers of B. thermosphacta (Fig. 4B and C). 3.2.3. Gram-negative bacteria After 4 and 6 weeks of storage, the number of Enterobacteriaceae was significantly ( P < 0.05) higher in treated samples than in untreated samples (Fig. 5A).
When chops were cut from the loins, the growth rate of Enterobacteriaceae on both the untreated and treated samples was similar, but the number of these organisms on treated samples was higher at all sampling times and significantly higher at some sampling times (Fig. 5B and C). There was no effect of treatment on the pseudomonad numbers on stored loins or on chops cut from them (data not shown).
4. Discussion Although pH was higher during storage, immediately after the loins had been treated, their pH was only 0.1 pH unit higher than on untreated loins. It is likely, therefore, that the difference in pH was due to the different bacterial population that grew on treated samples during storage. The number of LAB and AcLAB was lower at all times on treated samples and products of their metabolism would be less than in untreated samples. When the off-odour intensity and odour acceptability were evaluated, the chops that were treated with antimicrobial became unacceptable earlier than untreated chops. Comments from the panelists suggest that the odour was putrid and quite different on the treated samples than the off-odours usually encountered. Among the bacterial groups enumerated, the only one that had higher numbers in treated samples were the Enterobacteriaceae and their numbers were
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
only about 0.5 log units higher in treated samples. Usually this group is not considered to be a major contributor to aerobic spoilage if glucose has not been exhausted (Gill, 1986). On loins, after 4 and 6 weeks of storage in vacuum, the number of Enterobacteriaceae was about 1 log unit higher on treated loins than on untreated loins. No perceptible difference in offodour intensity and odour acceptance was noted between the samples. It was only upon transfer of the samples to an aerobic environment and a higher temperature that differences in odour between treated and untreated samples were perceived. There is little information about the interaction of different organisms upon transfer from an anoxic environment to an aerobic one, particularly when the meat has been treated with an effective antimicrobial. When pork was transferred from an environment of 100% carbon dioxide to retail display, the LAB remained the dominant flora during retail display and growth of gram-negative bacteria was delayed (Greer et al., 1993). In this work, there was also a generation of off-odours which was not easily explained. A very detailed analysis of the bacterial population present when off-odours become apparent would be required to properly address this question. Early in these experiments, the data suggested that, except for an apparent increase in the population of Enterobacteriaceae, there was no effect of the antimicrobial mixture. When colonies on MRS agar were carefully observed, colonial differences were apparent, even though the total number of colonies on plates from treated and untreated samples was approximately the same. A further evaluation of the colonies demonstrated that, in the treated samples, the majority of the colonies on MRS agar were gram-negative and catalase positive, thus masking the efficacy of the antimicrobial against the LAB. Therefore, all colonies on MRS agar were differentiated using the catalase test and results were interpreted based on this reaction. When semi-selective media such as MRS or nonselective media are used in studies such as this one where the ‘‘natural’’ balance of the bacterial population is disrupted, it is possible that the effect of the antimicrobial treatment could be missed without careful observation of results. The enumeration of AcLAB whose numbers are usually 1 to 2 log units/cm2 lower than LAB enumerated on MRS confirms the efficacy of the antimicrobial and the difficulty with consider-
265
ing total counts on MRS agar to be those of presumptive LAB under these experimental conditions. The antimicrobial treatment evaluated controlled the growth of LAB and B. thermosphacta during 6 weeks of storage in vacuum. Treated samples had lower numbers of bacteria, which were not only statistically significant but also were greater than 1 log lower which would have practical significance as well. It is unusual for an antimicrobial treatment to inhibit these groups of bacteria for an extended storage period. When steaks were cut from loins, their initial numbers reflected the microbial condition of the loins. If the differences in numbers between untreated and treated samples were large, the differences on the steaks were greater than when the numbers on treated and untreated samples were more similar. During increased time in retail display, the numbers on treated samples generally increased to reach the same levels as untreated samples after 7 days. From a practical point of view, the effect of the antimicrobial combination was more pronounced during anoxic storage than during retail display. Most studies have been conducted on pure cultures and even with controlled conditions the efficacy of antimicrobials reduces significantly with increased storage time (Cutter and Siragusa, 1994, 1996a,b, 1997; El-Khateib et al., 1993). In previous work with the 3:1 mixture of lysozyme and nisin on sterile pork and pure bacterial cultures (Nattress et al., 2001), synergy between nisin and lysozyme was demonstrated. It was hypothesized that nisin had an immediate effect on the bacteria and that the antimicrobial effect was extended during storage by lysozyme or the mixture of lysozyme and nisin. Clearly, the 3:1 mixture of lysozyme and nisin was effective at controlling the growth of a variety of LAB that would be found on naturally contaminated pork. This includes extremely good activity against the AcLAB which have been associated with the development of off-flavours in beef stored in 100% carbon dioxide (Nattress and Jeremiah, 2000). If the spectrum of activity of these antimicrobials could be extended to include the gram-negative microflora and if they do not affect the flavour of the meat, they could be used in fresh meat systems. The spectrum of activity could be increased by altering the structure of the lysozyme (Ibrahim et al., 1994; Masschalck et al., 2001) or by covalently modifiying it (Ibrahim et al., 1994). Chelators such as EDTA or lactoferrin
266
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267
could also be included in the antimicrobial mixture, although their activity in situ has not been promising (Cutter and Siragusa, 1995; Gill and Holley, 2000a). Alternatively, another antimicrobial such as lactate (Scannell et al., 1997), trisodium phosphate (Carneiro de Melo et al., 1998) or a polymerized amino acid (Shima et al., 1984; Ting et al., 1999) could be included in the mix that could potentially act synergistically with lysozyme and nisin. There has been considerable speculation and little data to describe what would happen to the microbial ecology when a group is inhibited. In this study, the inhibition of the LAB and the AcLAB appears to have resulted in a slight increase in the number of Enterobacteriaceae, and from the evaluation of odour attributes, possibly a shift in the population to strains that might cause early spoilage of the meat. Cudjoe (1988) reported a small increase in number of coliforms on cow heads that had been sprayed with lactic acid and stored in vacuum for 12 days when compared with unsprayed heads. Sensory evaluation did not differ between treated and untreated samples. Escherichia coli O157:H7 inoculated onto decontaminated and untreated beef grew to higher numbers on the decontaminated muscle than on the untreated muscle when samples were incubated for up to 15 days in vacuum at 10 jC, but the total number of bacteria was not different (Nissen et al., 2001). There is a need for a more detailed evaluation of the population shifts in the presence of this lysozyme/nisin antimicrobial mixture, particularly if an effort is undertaken to expand its spectrum of activity.
Acknowledgements Funding for this work was provided by Canadian Inovatech, Abbotsford, BC, Canada and the Matching Investment Initiative of Agriculture and Agri-Food Canada.
References Baird, R.M., Corry, J.E.L., Curtis, G.D.W. (Eds.), 1987. Pharmacopoeia of Culture Media for Food Microbiology. Int. J. Food Microbiol., vol. 5, pp. 187 – 299. Bester, B.H., Lombard, S.H., 1990. Influence of lysozyme on se-
lected bacteria associated with Gouda cheese. J. Food Prot. 53, 306 – 311. Carneiro de Melo, A.M., Cassar, C.A., Miles, R.J., 1998. Trisodium phosphate increases sensitivity of Gram-negative bacteria to lysozyme and nisin. J. Food Prot. 61, 839 – 843. Chung, W., Hancock, R.E., 2000. Action of lysozyme and nisin mixtures against lactic acid bacteria. Int. J. Food Microbiol. 60, 25 – 32. Chung, K.-T., Dickson, J.S., Crouse, J.D., 1989. Effects of nisin on growth of bacteria attached to meat. Appl. Environ. Microbiol. 55, 1329 – 1333. CIE (Commission Internationale de l’Eclairage), 1978. Recommendations on uniform color spaces – color difference equation – psychometric color terms. CIE publication no. 15 (E-1.3.1) 1891/(TC-1.3), supplement no. 2, Paris, France, pp. 8 – 12. Cudjoe, K.S., 1988. The effect of lactic acid sprays on the keeping qualities of meat during storage. Int. J. Food Microbiol. 7, 1 – 7. Cunningham, F.E., Proctor, V.A., Goetsch, S.J., 1991. Egg-white lysozyme as a food preservative: an overview. World’s Poult. Sci. J. 47, 141 – 163. Cutter, C.N., Siragusa, G.R., 1994. Decontamination of beef carcass tissue with nisin using a pilot scale model carcass washer. Food Microbiol. 11, 481 – 489. Cutter, C.N., Siragusa, G.R., 1995. Population reductions of Gramnegative pathogens following treatment with nisin and chelators under various conditions. J. Food Prot. 58, 977 – 983. Cutter, C.N., Siragusa, G.R., 1996a. Reductions of Listeria innocua and Brochothrix thermosphacta on beef following nisin spray treatments and vacuum packaging. Food Microbiol. 13, 23 – 33. Cutter, C.N., Siragusa, G.R., 1996b. Reduction of Brochothrix thermosphacta on beef surfaces following immobilization of nisin in calcium alginate gels. Lett. Appl. Microbiol. 23, 9 – 12. Cutter, C.N., Siragusa, G.R., 1997. Growth of Brochothrix thermosphacta in ground beef following treatments with nisin in calcium alginate gels. Food Microbiol. 14, 425 – 430. Cutter, C.N., Siragusa, G.R., 1998. Incorporation of nisin into a meat binding system to inhibit bacteria on beef surfaces. Lett. Appl. Microbiol. 27, 19 – 23. Davidson, P.M., Post, L.S., Branen, A.L., McCurdy, A.R., 1993. Naturally occurring and miscellaneous food antimicrobials. In: Branen, A.L., Davidson, P.M. (Eds.), Antimicrobials in Foods. Marcel Dekker, New York, pp. 371 – 419. Delves-Broughton, J., Blackburn, P., Evans, R.J., Hugenholtz, J., 1996. Applications of the bacteriocin nisin. Antonie van Leeuwenhoek 69, 193 – 202. During, K., Porsch, P., Mahn, A., Brinkmann, O., Gieffers, W., 1999. The non-enzymatic microbiocidal activity of lysozymes. FEBS Lett. 449, 93 – 100. El-Khateib, T., Yousef, A.E., Ockerman, H.W., 1993. Inactivation and attachment of Listeria monocytogenes on beef muscle treated with lactic acid and selected bacteriocins. J. Food Prot. 56, 29 – 33. Gill, C.O., 1986. The control of microbial spoilage in fresh meats. In: Pearson, A.M., Dutson, T.R. (Eds.), Advances in Meat Research, vol. 2. AVI Publishing, New York, pp. 49 – 88. Gill, A.O., Holley, R.A., 2000a. Inhibition of bacterial growth on
F.M. Nattress, L.P. Baker / International Journal of Food Microbiology 85 (2003) 259–267 ham and bologna by lysozyme, nisin and EDTA. Food Res. Int. 33, 83 – 91. Gill, A.O., Holley, R.A., 2000b. Surface application of lysozyme, nisin, and EDTA to inhibit spoilage and pathogenic bacteria on ham and bologna. J. Food Prot. 63, 1338 – 1346. Greer, G.G., Jeremiah, L.E., 1980. Influence of retail display temperature on psychrotrophic bacterial growth and beef case-life. J. Food Prot. 34, 542 – 546. Greer, G.G., Jones, S.D.M., 1996. Quality and bacteriological consequences of beef carcass spray-chilling: effects of spray duration and boxed beef storage temperature. Meat Sci. 45, 61 – 73. Greer, G.G., Dilts, B.D., Jeremiah, L.E., 1993. Bacteriology and retail case life of pork after storage in carbon dioxide. J. Food Prot. 56, 689 – 693. Holzapfel, W.H., Geisen, R., Schillinger, U., 1995. Biological preservation of foods with reference to protective cultures, bacteriocins and food-grade enzymes. Int. J. Food Microbiol. 24, 343 – 362. Ibrahim, H.R., Hatta, H., Fujiki, M., Kim, M., Yamamoto, T., 1994. Enhanced antimicrobial action of lysozyme against Gram-negative and Gram-positive bacteria due to modification with preillaldehyde. J. Agric. Food Chem. 42, 1813 – 1817. Ibrahim, H.R., Higashiguchi, S., Koketsu, M., Juneja, L.R., Kim, M., Yamamoto, T., Sugimoto, Y., Aoki, T., 1996. Partially unfolded lysozyme at neutral pH agglutinates and kills Gram-negative and Gram-positive bacteria through membrane damage mechanism. J. Agric. Food Chem. 44, 3799 – 3806. Masschalck, B., Van Houdt, R., Van Haver, E.G., Michiels, C.W., 2001. Inactivation of gram-negative bacteria by lysozyme, denatured lysozyme, and lysozyme-derived peptides under high hydrostatic pressure. Appl. Environ. Microbiol. 67, 339 – 344.
267
Nattress, F.M., Jeremiah, L.E., 2000. Bacterial mediated off-flavours in retail-ready beef after storage in controlled atmospheres. Food Res. Int. 33, 743 – 748. Nattress, F.M., Yost, C.K., Baker, L.P., 2001. Evaluation of the ability of lysozyme and nisin to control meat spoilage bacteria. Int. J. Food Microbiol. 70, 111 – 119. Nissen, H., Maugesten, T., Lea, P., 2001. Survival and growth of Escherichia coli O157:H7, Yersinia enterocolitica and Salmonella enteritidis on decontaminated and untreated meat. Meat Sci. 57, 291 – 298. Padgett, T., Han, I.Y., Dawson, P.L., 1998. Incorporation of foodgrade antimicrobial compounds into biodegradable packaging films. J. Food Prot. 61, 1330 – 1335. Rose, N.L., Sporns, P., Stiles, M.E., McMullen, L.M., 1999. Inactivation of nisin by glutathione in fresh meat. J. Food Sci. 64, 759 – 762. SAS Institute, 1995. SAS/STAT User’s Guide. SAS Institute, Cary, NC. Scannell, A.G.M., Hill, C., Buckely, D.J., Arendt, E.K., 1997. Determination of the influence or organic acids and nisin on shelflife and microbiological safety aspects of fresh pork sausage. J. Appl. Microbiol. 83, 407 – 412. Shima, S., Matsuoka, H., Iwamoto, T., Sakai, H., 1984. Antimicrobial action of a-poly-L-lysine. J. Antibiot. 37, 1449 – 1455. Ting, H.Y., Ishizaki, S., Tanaka, M., 1999. a-Polylysine improves the quality of surimi products. J. Muscle Foods 10, 279 – 294. Zhang, S., Mustapha, A., 1999. Reduction of Listeria monocytogenes and Escherichia coli O157:H7 numbers on vacuum-packaged fresh beef treated with nisin or nisin combined with EDTA. J. Food Prot. 62, 1123 – 1127.