De invloed van de (micro)structuur van levensmiddelen op degroeikinetiek van Listeria monocytogenes.Toepassing op vleeswaren

Most available predictive models are developed on the basis of data obtained in liquid microbiological media, or by challenge tests in real foods. Although liquid microbiological media are convenient and reproducible, they do not include the effect of different food (micro)structures, such as aqueous gels, emulsions and gelled emulsions. On the other hand, in challenge tests all factors influencing microbial growth are included, but they lack in reproducibility, are expensive and time-consuming. The aim of this dissertation was to investigate the effect of the food matrix complexity on the growth dynamics of L. monocytogenes, with particular focus on the influence of (i) the background microflora and (ii) the physicochemical characteristics and processing technology on Frankfurter sausages and (iii) the food (micro)structure by means of food (model) systems. The target microorganism was Listeria monocytogenes, as it is one of the major microbial threats in the food industry. The target food system was Frankfurter sausages, as they are easily available in market, complex in (micro)structure and related to cases of listeriosis. Model systems, which cover the whole spectrum of (micro)structural complexity and at the same time approach the compositional and physicochemical characteristics of the target food system, were produced. Firstly, it was necessary to define the fastest growing strain of L. monocytogenes, among three isolates from meat products. Growth dynamics of L. monocytogenes strains were compared with the aid of the Time-To-Detection (TTD) method, under physicochemical conditions similar to Frankfurters. This method exhibited a few limitations at stressing conditions and low inoculum levels. A sensitivity analysis of the factors that influence the output of this method (growth parameters) was carried out. Results demonstrated that the output of the method was significantly affected by all factors. The factors exhibiting the highest influence on the results were (i) the calibration curve relating viable plate counts and Optical Density (OD) data; (ii) the approach to estimate TTD values; (iii) the detection limit of OD measurements; and (iv) the range of the initial cell concentrations considered (N0). Overall, lag phase (λ) estimates were more sensitive than maximum specific growth rate (μmax) estimates. The approach to estimate TTD values and the OD detection limit were the most influential factors for the μmax and λ estimation. Secondly, the effect of the dominant indigenous microflora (Leuconostoc carnosum) was studied on L. monocytogenes growth on vacuum packed Frankfurter sausages. Results demonstrated that at low temperature and high L. carnosum inoculum level, L. monocytogenes stops growing earlier than L. carnosum and the lowest maximum population is reached. Two models were used to fit the experimental data in order to facilitate the interpretation of the results. The Dens et al. model described species interactions, which revealed a predominant effect of L. carnosum on L. monocytogenes, and described the decrease phase of L. monocytogenes population. The Baranyi and Roberts model, a special case of the Dens et al. model for monoculture, estimated the apparent maximum population and brought into question the validity of the Jameson effect at low temperature, which describes the simultaneous halt of growth of two species in co-culture. Furthermore, compositional, physicochemical and processing aspects were evaluated on their influence on the growth dynamics of L. monocytogenes on real vacuum-packed Frankfurters. This work demonstrated that even products of the same commercial type might exhibit significant differences on their growth dynamics of L. monocytogenes. Results indicated that there were no major significant differences in physicochemical characteristics for two types of sausages but, there was statistically significant variability at the concentration of preservatives. Liquid systems, with the same preservatives’ level as the two types of sausages, exhibited a lower variability in growth dynamics of L. monocytogenes than the two types of sausages. This observation led to the conclusion that other factors, such as the presence or absence of the sausage casing, and differences in the (micro)structure and processing factors affect the growth dynamics ofL. monocytogenes on different types of Frankfurter sausages. The previous results served as a benchmark for the design and production of the model systems. The growth dynamics of L. monocytogenes were studied in/on these food (model) systems (liquids, aqueous gels, emulsions, gelled emulsions and canned meat, which was specifically ordered for the needs of this study and had specified composition). Results showed that L. monocytogenes grows faster on canned meat, then in the liquids followed by aqueous gels, gelled emulsions and finally in emulsions. These findings emphasised that predictive models that do not include the effect of food (micro)structure might underestimate the growth of foodborne pathogens, which poses a risk for the microbiological food safety. The importance of developing suitable food (model) systems that approach real food products and isolate the effect of the (micro)structure has been introduced in this dissertation, as it still remains a challenge. Overall, experimental data need to be collected and predictive models, which incorporate the combination of all factors influencing microbial growth need to be developed, in order to contribute and ensure microbiological food safety.status: publishe