Biopreservation in modified atmosphere packaged vegetables

Recent trends in food preservation are the use of mild preservation techniques, such as modified atmosphere (MA) packaging and refrigeration, to prolong the shelflife of foods without affecting the fresh character of the product. This has resulted in the development of a new generation of chill stored, minimally processed foods, such as vegetables that are packaged under reduced oxygen (0 2 ) and elevated carbon dioxide (C0 2 ) concentrations. These gas atmospheres in combination with refrigerated storage conditions reduce the respiration rates of these products, and can limit the growth of endogenous spoilage bacteria, thereby rendering prolonged shelflives as compared with storage under ambient conditions. Although MA packaged vegetables have a fair safety record, they may incidently harbor foodborne pathogens. Especially the outgrowth of psychrotrophic (cold tolerant) pathogens can be considered a safety concern, but quantitative data on their growth under MA conditions that are suitable for produce are essentially lacking. Detailed information on the outgrowth of both pathogens and vegetable-associated spoilage bacteria under these conditions may give better insight into possible antagonistic action of competing microflora on the growth of pathogens, and the development of adequate countermeasures toward their growth to hazardous levels.One of the objectives of the studies that are described in this thesis was to obtain information on the impact of refrigerated MA storage conditions on the growth of microorganisms on minimally processed, MA packaged produce. Since our initial studies substantiated the possible hazard which can be posed by psychrotrophic pathogens, the use of biopreservation for adequate control of these microorganisms was investigated. In this respect, we have focused on lactic acid bacteria (LAB) because they occur naturally on fresh and minimally processed vegetables, and are able to produce a variety of antimicrobial substances, amongst which bacteriocins.Chapter 1 gives a short introduction on the technology of MA packaging of minimally processed vegetables and their current microbiological safety and quality status. Furthermore, the possible role of LAB as biopreservation agents in foods is addressed with special emphasis on those strains that are able to produce bacteriocins. The characteristics and mode of action of different bacteriocins that are produced by LAB are described, as well as aspects which are relevant to the application of these antimicrobial compounds or their producing organisms as food preservatives.Detailed information on the impact of refrigerated MA storage on the population dynamics of the endogenous microflora on vegetables is limited. Therefore, the prevailing epiphytic microorganisms on model produce ( i.e . chicory endive, and mungbean sprouts) were identified before and after MA storage. Subsequently, the growth characteristics of the predominant spoilage bacteria were determined under MA conditions ( Chapter 2 ). It was found that the major strains on both products were Enterobacteriaceae and Pseudomonas species. There were changes in the prevalent species on chicory endive before and after MA storage, but this was not observed for mungbean sprouts. To quantify the influence of C0 2 and 0 2 on the growth of the individual spoilage bacteria, a model agar system was developed that mimicked the in situ growth of bacteria on the surface of vegetables. The analysis of the growth of the predominant strains under controlled 0 2 and C0 2 concentrations at 8°C indicated that, in general, lag times were not present under the various conditions tested, and that the maximum population sizes were not affected. However, the maximum specific growth rates generally decreased with increasing C0 2 concentrations, which was independent of the 0 2 concentrations applied. This effect was more pronounced for Pseudomonas species than for Enterobacteriaceae. Representatives of the former subpopulation already showed significantly reduced maximum specific growth rates at C0 2 concentrations that are suitable for MA packaged vegetables.In Chapter 3 , the effects of different 0 2 and C0 2 concentrations on the growth of Aeromonas hydrophila, Yersinia enterocolitica, L. monocytogenes, and a cold tolerant strain of Bacillus cereus were quantified in the above mentioned model agar system at 8°C. Again, the maximum specific growth rates of the various microorganisms decreased significantly with increasing C0 2 concentration, while the maximum population densities were not affected and lag times were not observed. Extrapolation of the results from the model system to MA packaged vegetables suggested that growth of the above pathogens may occur at 8°C to the same maximum population densities as compared to ambient conditions, but at 10 to 20% lower maximum specific growth rates. Because we have shown that MA packaging can only moderately retard the growth of psychrotrophic pathogens, these bacteria can be considered a safety hazard to which countermeasures need to be developed.L. monocytogenes is a relevant psychrotrophic pathogen on minimally processed vegetables that requires adequate control measures. This Gram-positive bacterium can be inhibited by a number of bacteriocins that are produced by LAB. Thus, biopreservation using bacteriocin producing LAB strains may effectively prevent the growth of this bacterium, and possibly other psychrotrophic pathogens. With respect to practical application, an effort was made to obtain bacteriocinogenic LAB from minimally processed vegetables since these strains might be best adapted to the specific conditions under which their antimicrobial activity is required. Out of a total of 890 LAB isolates from mungbean sprouts and chicory endive, only nine strains were found to produce bacteriocins ( Chapter 4 ). Three of these strains exhibited antimicrobial activity towards a wide variety of Gram- positive bacteria, including the foodborne pathogens L.monocytogenes, and nonproteolytic Clostridium botulinum. Two of these bacteriocinogenic isolates were identified as Pediococcus parvulus on the basis of fermentation patterns and ribotyping. Recombinant DNA experiments revealed the presence of the pediocin PA-1 gene in both LAB strains, and further biochemical analysis confirmed the production of pediocin PA-1 in absence of other bacteriocins.The third broad spectrum bacteriocin-producing strain was identified as Enterococcus mundtii ( Chapter 5 ). Its bacteriocin was purified to homogeneity and characterized. Elucidation of the complete primary amino acid sequence revealed a novel bacteriocin of 43 amino acids (M r 4287.2), designated as mundticin, that belongs to the class IIa bacteriocins of LAB. Other LAB that produce this class of antimicrobial peptides have previously been isolated from a wide variety of meat and dairy products, and their apparent wide-spread occurrence in foods may allow for the selection of strains that are ecologically adapted to specific food environments.As further described in Chapter 5 , mundticin exhibited bactericidal activity towards L.monocytogenes, and was shown to dissipate the membrane potential of whole cells, and to deplete their intracellular ATP pools. Mundticin was selected as a prototype of class IIa bacteriocins to investigate the biophysical properties of this peptide by molecular computer analysis. It was found that mundticin and related bacteriocins, contain a central α-helical region that was predicted to insert at an angle of 30 to 50° in a simulated hydrophobic/hydrophilic interface. This oblique insertion is proposed to cause a destabilisation of the phospholipid bilayer and facilitate the insertion and/or aggregation of monomers into functional pores in the cytoplasmic membrane. This concept is of interest for the rational design of bacteriocins with, for example, better biopreservation properties.For the effective application of bacteriocins as biopreservation agents, it is essential to gain more insight into the variation in the susceptibility of Gram-positive bacteriocins towards bacteriocins. The comprehensive study that is described in Chapter 6 addresses several factors which are associated with the natural variability in the bacteriocin-sensitivity of closely related non- bacteriocinogenic LAB. Two sets of strains for which the minimal inhibitory concentrations for nisin and pediocin PA-1 differed 100 to 1000-fold were compared. Our data provide evidence that the association of bacteriocins with the cell membrane and their subsequent insertion into the membrane take place in a similar way for cells that have a high or a low natural bacteriocin tolerance. For insensitive strains, the overall constitution of the membrane, rather than the mere membrane fluidity, may preclude the formation of pores with sufficient diameters and lifetimes to ultimately cause cell death. Further analysis of this phenomena may require more attention for the structural role of proteins in biological membranes.The last part of this dissertation comprises a study on the potential of the vegetable- associated bacteriocin-producing bacteria and mundticin to act as biopreservation agents on refrigerated, MA-stored vegetables ( Chapter 7 ). Both P. parvulus strains were found to be unsuitable for this purpose, due to the lack of bacteriocin production at 4 and 8°C. By contrast, E. mundtii displayed a number of characteristics that favor an application as a biopreservative agent, such as limited acid production, production of a bacteriocin with anti-listeria activity at low temperatures, and high growth rates under MA conditions. Despite these properties, this strain failed to effectively inhibit the growth of L. monocytogenes on fresh mungbean sprouts. This is probably not due to a lack of in situ bacteriocin production, but rather the result of proteolytic degradation of mundticin or the growth of L. monocytogenes cells that are resistant to mundticin. Experiments with mundticin in a washing step and as a component of a coating were successful and warrant further research on the potential of this compound as a natural preservative agent for minimally processed vegetables.In conclusion, this dissertation has contributed to a better understanding of the microbiological characteristics of minimally processed vegetables that are stored under MA conditions. A general concern for the possible outgrowth of psychrotrophic pathogens on MA stored vegetables was substantiated by our investigations, which stresses that hygienic cultivation and processing conditions of this type of products remain indispensable. Although bacteriocins cannot be used as the sole preservatives to enhance the safety of refrigerated MA-stored minimally processed vegetables, they can contribute to tackle safety problems that may arise from certain cold tolerant, Grampositive pathogens