MAP, product safety and nutritional quality F. Devlieghere and J. Debevere, Ghent University, Belgium and M I Gil, CEBAS-CSIC, Spain 11.1 Introduction Modified atmosphere packaging(MAP) may be defined as the enclosure of food products in gas-barrier materials, in which the gaseous environment has been changed(Young et al, 1988). Because of its substantial shelf-life extending effect, MAP has been one of the most significant and innovative growth areas in retail food packaging. The potential advantages and disadvantages of mAP have been presented by Farber (1991), Parry (1993) and Davies(1995) Whilst there is considerable information available regarding suitable mixtures for different food products, there is still a lack of scientific detail regarding many aspects relating to MAP. These include mechanism of action of carbon dioxide(CO2)on microorganisms safety of MAP packaged food products effect of MAP on the nutritional quality of packaged food products Current research and gaps in knowledge are discussed in the following sections 11.2 Carbon dioxide as an antimicrobial gas The gases that are applied in MAP today are basically O2, CO2 and N2. The last has no specific preservative effect but functions mainly as a filler gas to avoid the collapse that takes place when CO2 dissolves in the food product. CO because of its antimicrobial actrviduyced into the package, it is partly dissolved in
11.1 Introduction Modified atmosphere packaging (MAP) may be defined as ‘the enclosure of food products in gas-barrier materials, in which the gaseous environment has been changed’ (Young et al., 1988). Because of its substantial shelf-life extending effect, MAP has been one of the most significant and innovative growth areas in retail food packaging. The potential advantages and disadvantages of MAP have been presented by Farber (1991), Parry (1993) and Davies (1995). Whilst there is considerable information available regarding suitable gas mixtures for different food products, there is still a lack of scientific detail regarding many aspects relating to MAP. These include: • mechanism of action of carbon dioxide (CO2) on microorganisms • safety of MAP packaged food products • effect of MAP on the nutritional quality of packaged food products. Current research and gaps in knowledge are discussed in the following sections. 11.2 Carbon dioxide as an antimicrobial gas The gases that are applied in MAP today are basically O2, CO2 and N2. The last has no specific preservative effect but functions mainly as a filler gas to avoid the collapse that takes place when CO2 dissolves in the food product. CO2, because of its antimicrobial activity, is the most important component in applied gas mixtures. When CO2 is introduced into the package, it is partly dissolved in 11 MAP, product safety and nutritional quality F. Devlieghere and J. Debevere, Ghent University, Belgium and M I Gil, CEBAS-CSIC, Spain
MAP, product safety and nutritional quality 209 the water phase and the fat phase of the food. This results, after equilibrium, in a certain concentration of dissolved CO2([CO2ldiss )in the water phase of the product. Devlieghere et al. (1998)have demonstrated that the growth inhibition of microorganisms in modified atmospheres is determined by the concentration of dissolved cO, in the water phase The effect of the gaseous environment on microorganisms in foods is not as well understood by microbiologists and food technologists as are other external factors, such as pH and aw. Despite numerous reports of the effects of COz on microbial growth and metabolism, the 'mechanism' of CO2 inhibition still remains unclear (Dixon and Kell, 1989; Day, 2000). The question of whether any specific metabolic pathway or cellular activity is critically sensitive to CO inhibition has been examined in several studies. The different proposed mechanisms of action are. 1. Lowering the pH of the food 2. Cellular penetration followed by a decrease in the cytoplasmic pH of the 3. Specific actions on cytoplasmic enzymes 4. Specific actions on biological membranes When gaseous CO2 is applied to a biological tissue, it first dissolves in the iquid phase, where hydration and dissociation lead to a rapid pH decrease in the tissue. This drop in pH, which depends on the buffering capacity of the medium Dixon and Kell, 1989), is not large in food products. In fact, the pH drop in cooked meat products only amounted to 0.3 pH units when 80% of CO2 was applied in the gas phase with a gas/product volume ration of 4: 1 Devlieghere et al., 2000b). Several studies have proved that the observed inhibitory effects of COz could not solely be explained by the acidification of the substrate( Becker, 1933 Coyne,1933) Many researchers have documented the rapidity with which CO2 in solution penetrates into the cell. Krogh(1919) discovered that this rate is 30 times faster than for oxygen(O2) under most circumstances. Wolfe(1980) suggested the inhibitory effects of CO2 are the result of internal acidification of the cytoplasm Eklund(1984) supported this idea by pointing out that the growth inhibition of four bacteria obtained with CO2 had the same general form as that obtained with weak organic acids(chemical preservatives), such as sorbic and benzoic acid. Tan and Gill (1982)also found that the intracellular pH of Pseudomonas fluorescens fell by approximately 0.03 units for each I mM rise in extracellular CO2 concentration CO2 may also exert its influence upon a cell by affecting the rate at which particular enzymatic reactions proceed. One way this may be brought about is to cause an alteration in the production of a specific enzyme, or enzymes, via induction or repression of enzyme synthesis(Dixon, 1988; Dixon and Kell 1989, Jones, 1989 ). It was also suggested ones and greenfield, 1982, Dixon and Kell, 1989)that the primary sites where CO2 exerts its effects are the enzymatic carboxylation and decarboxylation reactions, although inhibition of other enzymes has also been reported (Jones and Greenfield, 1982
the water phase and the fat phase of the food. This results, after equilibrium, in a certain concentration of dissolved CO2 ([CO2]diss) in the water phase of the product. Devlieghere et al. (1998) have demonstrated that the growth inhibition of microorganisms in modified atmospheres is determined by the concentration of dissolved CO2 in the water phase. The effect of the gaseous environment on microorganisms in foods is not as well understood by microbiologists and food technologists as are other external factors, such as pH and aw. Despite numerous reports of the effects of CO2 on microbial growth and metabolism, the ‘mechanism’ of CO2 inhibition still remains unclear (Dixon and Kell, 1989; Day, 2000). The question of whether any specific metabolic pathway or cellular activity is critically sensitive to CO2 inhibition has been examined in several studies. The different proposed mechanisms of action are: 1. Lowering the pH of the food. 2. Cellular penetration followed by a decrease in the cytoplasmic pH of the cell. 3. Specific actions on cytoplasmic enzymes. 4. Specific actions on biological membranes. When gaseous CO2 is applied to a biological tissue, it first dissolves in the liquid phase, where hydration and dissociation lead to a rapid pH decrease in the tissue. This drop in pH, which depends on the buffering capacity of the medium (Dixon and Kell, 1989), is not large in food products. In fact, the pH drop in cooked meat products only amounted to 0.3 pH units when 80% of CO2 was applied in the gas phase with a gas/product volume ration of 4:1 (Devlieghere et al., 2000b). Several studies have proved that the observed inhibitory effects of CO2 could not solely be explained by the acidification of the substrate (Becker, 1933; Coyne, 1933). Many researchers have documented the rapidity with which CO2 in solution penetrates into the cell. Krogh (1919) discovered that this rate is 30 times faster than for oxygen (O2) under most circumstances. Wolfe (1980) suggested the inhibitory effects of CO2 are the result of internal acidification of the cytoplasm. Eklund (1984) supported this idea by pointing out that the growth inhibition of four bacteria obtained with CO2 had the same general form as that obtained with weak organic acids (chemical preservatives), such as sorbic and benzoic acid. Tan and Gill (1982) also found that the intracellular pH of Pseudomonas fluorescens fell by approximately 0.03 units for each 1 mM rise in extracellular CO2 concentration. CO2 may also exert its influence upon a cell by affecting the rate at which particular enzymatic reactions proceed. One way this may be brought about is to cause an alteration in the production of a specific enzyme, or enzymes, via induction or repression of enzyme synthesis (Dixon, 1988; Dixon and Kell, 1989; Jones, 1989). It was also suggested (Jones and Greenfield, 1982; Dixon and Kell, 1989) that the primary sites where CO2 exerts its effects are the enzymatic carboxylation and decarboxylation reactions, although inhibition of other enzymes has also been reported (Jones and Greenfield, 1982). MAP, product safety and nutritional quality 209
210 Novel food packaging techniques Another possible factor contributing to the growth-inhibitory effect of Co could be an alteration of the membrane properties(Daniels et al, 1985, Dixon and Kell, 1989). It was suggested that CO2 interacts with lipids in the cell membrane, decreasing the ability of the cell wall to uptake various ions Moreover, perturbations in membrane fluidity, caused by the disordering of the lipid bilayer, are postulated to alter the function of membrane proteins( Chin et a.,1976;Roth,1980) Studies examining the effect of a CO2 enriched atmosphere on the growth of microorganisms are often difficult to compare because of the lack of information regarding the packaging configurations applied. The gas/product volume ratio and the permeability of the applied film for O and CO, will influence the mount of dissolved CO2 and thus the microbial inhibition of the atmosphere For this reason, the concentration of dissolved CO2 in the aqueous phase of the food should always be measured and mentioned in publications concerning MAP (Devlieghere et al, 1998) Only a few publications deal with the effect of MAP on specific spoilage microorganisms. Gill and Tan(1980) compared the effect of CO2 on the growth of some fresh meat spoilage bacteria at 30oC. Molin(1983)determined the resistance to COz of several food spoilage bacteria. Boskou and Debevere(1997 1998) investigated the effect of CO2 on the growth and trimethy lamine ocLc ion of Shewanella putrefaciens in marine fish, and Devlieghere and Debevere(2000) compared the sensitivity for dissolved CO2 of different ilage bacteria at 7C. In general, Gram-negative microorganisms such as Pseudomonas, Shewanella and Aeromonas are very sensitive to CO2. Gram positive bacteria show less sensitivity and lactic acid bacteria are the most resistant. Most yeasts and moulds are also sensitive to CO2. The effect of CO2 on psychrotrophic food pathogens is discussed in Section 11.3 11.3 The microbial safety of MAP: Clostridium botulinum and Listeria monocytogenes Modified atmospheres containing CO] are effective in extending the shelf-life of many food products. However, one major concern is the inhibition of normal aerobic spoilage bacteria and the possible growth of psychrotrophic food pathogens, which may result in the food becoming unsafe for consumption before it appears to be organoleptically unacceptable. Most of the pathogenic bacteria can be inhibited by low temperatures(<7C). At these conditions, only psychrotrophic pathogens can proliferate. The effect of CO2 on the different psychrotrophic foodborne pathogens is described below a particular concern is the possibility that psychrotrophic, non strains of C. botulinum types B, E and F are able to grow and prod under MAP conditions. Little is known about the effects of modified atmosphere storage conditions on toxin production by C. botulinum. The possibility of nhibiting C. botulinum by incorporating low levels of Oz in the package does
Another possible factor contributing to the growth-inhibitory effect of CO2 could be an alteration of the membrane properties (Daniels et al., 1985; Dixon and Kell, 1989). It was suggested that CO2 interacts with lipids in the cell membrane, decreasing the ability of the cell wall to uptake various ions. Moreover, perturbations in membrane fluidity, caused by the disordering of the lipid bilayer, are postulated to alter the function of membrane proteins (Chin et al., 1976; Roth, 1980). Studies examining the effect of a CO2 enriched atmosphere on the growth of microorganisms are often difficult to compare because of the lack of information regarding the packaging configurations applied. The gas/product volume ratio and the permeability of the applied film for O2 and CO2 will influence the amount of dissolved CO2 and thus the microbial inhibition of the atmosphere. For this reason, the concentration of dissolved CO2 in the aqueous phase of the food should always be measured and mentioned in publications concerning MAP (Devlieghere et al., 1998). Only a few publications deal with the effect of MAP on specific spoilage microorganisms. Gill and Tan (1980) compared the effect of CO2 on the growth of some fresh meat spoilage bacteria at 30ºC. Molin (1983) determined the resistance to CO2 of several food spoilage bacteria. Boskou and Debevere (1997; 1998) investigated the effect of CO2 on the growth and trimethylamine production of Shewanella putrefaciens in marine fish, and Devlieghere and Debevere (2000) compared the sensitivity for dissolved CO2 of different spoilage bacteria at 7ºC. In general, Gram-negative microorganisms such as Pseudomonas, Shewanella and Aeromonas are very sensitive to CO2. Grampositive bacteria show less sensitivity and lactic acid bacteria are the most resistant. Most yeasts and moulds are also sensitive to CO2. The effect of CO2 on psychrotrophic food pathogens is discussed in Section 11.3. 11.3 The microbial safety of MAP: Clostridium botulinum and Listeria monocytogenes Modified atmospheres containing CO2 are effective in extending the shelf-life of many food products. However, one major concern is the inhibition of normal aerobic spoilage bacteria and the possible growth of psychrotrophic food pathogens, which may result in the food becoming unsafe for consumption before it appears to be organoleptically unacceptable. Most of the pathogenic bacteria can be inhibited by low temperatures (<7ºC). At these conditions, only psychrotrophic pathogens can proliferate. The effect of CO2 on the different psychrotrophic foodborne pathogens is described below. A particular concern is the possibility that psychrotrophic, non-proteolytic strains of C. botulinum types B, E and F are able to grow and produce toxins under MAP conditions. Little is known about the effects of modified atmosphere storage conditions on toxin production by C. botulinum. The possibility of inhibiting C. botulinum by incorporating low levels of O2 in the package does 210 Novel food packaging techniques
MAP, product safety and nutritional quality 211 not appear to be feasible. Miller(1988, cited by Connor et al., 1989)reported that psychrotrophic strains of C. botulinum are able to produce toxins in an environment with up to 10%O2. Toxin production by C. botulinum type E,prior to spoilage, has been described in three types of fish, at O2 levels of 2% and 4% OConnor-Shaw and Reyes, 2000). Dufresne et al.(2000) also proposed that additional barriers, other than headspace O2 and film, need to be considered to ensure the safety of MAP trout fillets, particularly at moderate temperature The probability of one spore of non-proteolytic C. botulinum(types B, E and F) being toxicogenic in rock fish was outlined in a report by Ikawa and Genigeorgis(1987). The results showed that the toxigenicity was significantly affected(P<0.005)by temperature and storage time, but not by the used modified atmosphere (vacuum, 100%CO2, or 70% CO2/30% air). In Tilapia fillets, a modified atmosphere(75% CO2/25% N2), at 8C, delayed toxin formation by C. botulinum type E, from 17 to 40 days, when compared to vacuum packaged fillets(Reddy et al, 1996). Similar inhibiting effects were recorded for salmon fillets and catfish fillets, at 4C (Reddy et al., 1997a and 1997b). Toxin production from non-proteolytic C. botulinum type B spores was also retarded by a CO2 enriched atmosphere(30% Co2/70% N2)in cooked turkey at 4C but not at 10C nor at 15.C (Lawlor et al, 2000). Recent results in a study by gibson et al.(2000)also indicated that 100% CO2 slows the growth rate of C. botulinum, and that this inhibitory effect is further enhanced with appropriate NaCl concentrations and chilled temperatures Listeria monocytogenes is considered a psychrotrophic foodborne pathogen Growth is possible at 1oC (Varnam and Evans, 1991) and has even been reported at temperatures as low as-15C (Hudson et al, 1994 ) The growth of L. monocytogenes in food products, packaged under modified atmospheres, has of several, although in some cases contradicting, stu de fernando et al., 1995). In general, L. monocytogenes is not greatly inhibited by CO2 enriched atmospheres(zhao et al, 1992)although when combined with other factors such as low temperature, decreased water activity and the addition of Na lactate the inhibiting effect of CO2 is significant (Devlieghere et al 2001). Listeria growth in anaerobic CO2 enriched atmosphere has been demonstrated in lamb in an atmosphere of 50: 50 CO2/N2, at 5C (Nychas 1994): in frankfurter type sausages in atmospheres of distinct proportions of CO2/N2, at 4, 7 and 10oC(Kramer and Baumgart, 1992)and in pork in an atmosphere of 40: 60 CO2/N2, at 4C(Manu-Tawiah et al., 1993).However other authors have not detected growth in chicken anaerobically packaged in 30:70 CO /N2, at 6C(Hart et al., 1991); in 75: 25 CO et al, 1990)and at 4C in 100% CO, in raw minced meat(franco-abuin et al. 997)or in buffered tryptose broth (Szabo and Cahill, 1998). Several investigations demonstrated possible growth of L. monocytogenes on modified atmosphere packaged fresh-cut vegetables, although the results depended very much on the type of vegetables and the storage temperature( Berrang et al. 1989a, Beuchat and Brackett, 1990; Omary et al., 1993, Carlin et al, 1995
not appear to be feasible. Miller (1988, cited by Connor et al., 1989) reported that psychrotrophic strains of C. botulinum are able to produce toxins in an environment with up to 10% O2. Toxin production by C. botulinum type E, prior to spoilage, has been described in three types of fish, at O2 levels of 2% and 4% (O’Connor-Shaw and Reyes, 2000). Dufresne et al. (2000) also proposed that additional barriers, other than headspace O2 and film, need to be considered to ensure the safety of MAP trout fillets, particularly at moderate temperature abuse conditions. The probability of one spore of non-proteolytic C. botulinum (types B, E and F) being toxicogenic in rock fish was outlined in a report by Ikawa and Genigeorgis (1987). The results showed that the toxigenicity was significantly affected (P<0.005) by temperature and storage time, but not by the used modified atmosphere (vacuum, 100% CO2, or 70% CO2/30% air). In Tilapia fillets, a modified atmosphere (75% CO2/25% N2), at 8ºC, delayed toxin formation by C. botulinum type E, from 17 to 40 days, when compared to vacuum packaged fillets (Reddy et al., 1996). Similar inhibiting effects were recorded for salmon fillets and catfish fillets, at 4ºC (Reddy et al., 1997a and 1997b). Toxin production from non-proteolytic C. botulinum type B spores was also retarded by a CO2 enriched atmosphere (30% CO2/70% N2) in cooked turkey at 4ºC but not at 10ºC nor at 15ºC (Lawlor et al., 2000). Recent results in a study by Gibson et al. (2000) also indicated that 100% CO2 slows the growth rate of C. botulinum, and that this inhibitory effect is further enhanced with appropriate NaC1 concentrations and chilled temperatures. Listeria monocytogenes is considered a psychrotrophic foodborne pathogen. Growth is possible at 1ºC (Varnam and Evans, 1991) and has even been reported at temperatures as low as ÿ1.5C (Hudson et al., 1994). The growth of L. monocytogenes in food products, packaged under modified atmospheres, has been the focus of several, although in some cases contradicting, studies (Garcia de Fernando et al., 1995). In general, L. monocytogenes is not greatly inhibited by CO2 enriched atmospheres (Zhao et al., 1992) although when combined with other factors such as low temperature, decreased water activity and the addition of Na lactate the inhibiting effect of CO2 is significant (Devlieghere et al., 2001). Listeria growth in anaerobic CO2 enriched atmosphere has been demonstrated in lamb in an atmosphere of 50:50 CO2/N2, at 5ºC (Nychas, 1994); in frankfurter type sausages in atmospheres of distinct proportions of CO2/N2, at 4, 7 and 10ºC (Kra¨mer and Baumgart, 1992) and in pork in an atmosphere of 40:60 CO2/N2, at 4ºC (Manu-Tawiah et al., 1993). However, other authors have not detected growth in chicken anaerobically packaged in 30:70 CO2/N2, at 6ºC (Hart et al., 1991); in 75:25 CO2/N2 at 4ºC (Wimpfheimer et al., 1990) and at 4ºC in 100% CO2 in raw minced meat (Franco-Abuin et al., 1997) or in buffered tryptose broth (Szabo and Cahill, 1998). Several investigations demonstrated possible growth of L. monocytogenes on modified atmosphere packaged fresh-cut vegetables, although the results depended very much on the type of vegetables and the storage temperature (Berrang et al., 1989a; Beuchat and Brackett, 1990; Omary et al., 1993; Carlin et al., 1995; MAP, product safety and nutritional quality 211
212 Novel food packaging techniques Carlin et al, 1996a and 1996b, Zhang and Farber, 1996, Juneja et al., 1998; Bennick et al, 1999, Jacxsens et al, 1999, Liao and Sapers, 1999: Thomas et 1., 1999; Castillejo-Rodriguez et al, 2000) There is no agreement about the effect of incorporating O2 in the atmosphere on the antimicrobial activity of CO2 on L. monocytogenes( Garcia de Fernando et al, 1995). However, this effect could be very important in practice, as the existence of residual O2 levels after packaging, and the diffusion of o2 through the packaging film, can result in substantial Oz levels during the storage of industrially anaerobically'modified atmosphere packaged food products. Most publications suggest there is a decrease in the inhibitory effect of CO2 on L monocytogenes when O2 is incorporated into the atmosphere. Experiments on raw chicken showed L. monocytogenes failed to grow at 4, 10 and 27 C, in an anaerobic atmosphere containing 75%CO2 and 25% N2(Wimpfheimer et al. 1990). However, an aerobic atmosphere containing 72.5%CO2, 22. 5% N2, and 5%O2 did not inhibit the growth of L. monocytogenes, even at 4C. L. monocytogenes was also only minimally inhibited on chicken legs, in an atmosphere containing 10% O2 and 90% CO2(Zeitoun and Debevere, 1991) There was no significant difference in the inhibitory effect of COz, between the range of o% and 50%, when 1.5%O2, or 21%O2 was present in the atmosphere of gas packaged brain heart infusion agar plates(Bennik et al. 1995). When L. monocytogenes was cultured in buffered nutrient broth, at 7.5C, in atmospheres containing 30%CO2, with four different O2 concentrations(0, 10, 20 and 40%) the results showed that bacterial growth increased with the increasing O concentrations(Hendricks and Hotchkiss, 1997) 11.4 The microbial safety of MAP: Yersinia enterocolitica and Aeromonas spp. Yersinia enterocolitica is generally regarded as one of the most psychrotrophic foodborne pathogens. Growth of y. enterocolitica was reported in vacuum packaged lamb at 0oC(Doherty et al, 1995, Sheridan and Doherty, 1994; Sheridan et aL., 1992), beef at -2C( Gill and Reichel, 1989), pork at 4C Bodnaruk and Draughon, 1998, Manu-Tawiah et al., 1993), fresh chicken breasts(Ozbas et al, 1997)and roast beef at 3C but not at -1 5C(Hudson et a.,1994) CO2 retards the growth of y. enterocolitica at refrigerated temperatures. The ffect of CO2 on the growth of Y. enterocolitica has been described by several luthors. Some of the results are shown in Table 11. 1. Oxygen also seems to play an inhibiting role on the growth of Y. enterocolitica( Garcia de fernando et al. 1995). To ensure total inhibition of Y. enterocolitica in O2 poor atmospheres and at realistic temperatures throughout the cooling chain, high CO2 concentrations Aeromonas species are able to multiply in food products stored in refrigerated conditions. Growth of A. hydrophila has been detected at low temperatures in a
Carlin et al., 1996a and 1996b; Zhang and Farber, 1996; Juneja et al., 1998; Bennick et al., 1999; Jacxsens et al., 1999; Liao and Sapers, 1999; Thomas et al., 1999; Castillejo-Rodriguez et al., 2000). There is no agreement about the effect of incorporating O2 in the atmosphere on the antimicrobial activity of CO2 on L. monocytogenes (Garcia de Fernando et al., 1995). However, this effect could be very important in practice, as the existence of residual O2 levels after packaging, and the diffusion of O2 through the packaging film, can result in substantial O2 levels during the storage of industrially ‘anaerobically’ modified atmosphere packaged food products. Most publications suggest there is a decrease in the inhibitory effect of CO2 on L. monocytogenes when O2 is incorporated into the atmosphere. Experiments on raw chicken showed L. monocytogenes failed to grow at 4, 10 and 27ºC, in an anaerobic atmosphere containing 75% CO2 and 25% N2 (Wimpfheimer et al., 1990). However, an aerobic atmosphere containing 72.5% CO2, 22.5% N2, and 5% O2 did not inhibit the growth of L. monocytogenes, even at 4ºC. L. monocytogenes was also only minimally inhibited on chicken legs, in an atmosphere containing 10% O2 and 90% CO2 (Zeitoun and Debevere, 1991). There was no significant difference in the inhibitory effect of CO2, between the range of 0% and 50%, when 1.5% O2, or 21% O2 was present in the atmosphere of gas packaged brain heart infusion agar plates (Bennik et al. 1995). When L. monocytogenes was cultured in buffered nutrient broth, at 7.5ºC, in atmospheres containing 30% CO2, with four different O2 concentrations (0, 10, 20 and 40%), the results showed that bacterial growth increased with the increasing O2 concentrations (Hendricks and Hotchkiss, 1997). 11.4 The microbial safety of MAP: Yersinia enterocolitica and Aeromonas spp. Yersinia enterocolitica is generally regarded as one of the most psychrotrophic foodborne pathogens. Growth of Y. enterocolitica was reported in vacuum packaged lamb at 0ºC (Doherty et al., 1995; Sheridan and Doherty, 1994; Sheridan et al., 1992), beef at ÿ2ºC (Gill and Reichel, 1989), pork at 4ºC (Bodnaruk and Draughon, 1998; Manu-Tawiah et al., 1993), fresh chicken breasts (O¨ zbas et al., 1997) and roast beef at 3ºC but not at ÿ1.5ºC (Hudson et al., 1994). CO2 retards the growth of Y. enterocolitica at refrigerated temperatures. The effect of CO2 on the growth of Y. enterocolitica has been described by several authors. Some of the results are shown in Table 11.1. Oxygen also seems to play an inhibiting role on the growth of Y. enterocolitica (Garcia de Fernando et al., 1995). To ensure total inhibition of Y. enterocolitica in O2 poor atmospheres and at realistic temperatures throughout the cooling chain, high CO2 concentrations in the headspace are necessary. Aeromonas species are able to multiply in food products stored in refrigerated conditions. Growth of A. hydrophila has been detected at low temperatures in a 212 Novel food packaging techniques