Inactivation of foodborne pathogen in ready-to-eat probiotic artichokes. First International Conference on Microbial Diversity

The present study aimed to follow the behaviour (survival/inactivation) of 3 foodborne pathogens Listeria monocytogenes ATCC19115, Salmonella enterica subsp. enterica ATCC13311 and Escherichia coli ATCC8739 in ready to eat (RTE) artichoke product (Standard) prepared by using an acidified brine and in that obtained by employing a mild fermentation step operated by the probiotic Lactobacillus paracasei strain LMGP22043 (Probiotic). Although the safety of the Standard product is regularly ascertained by microbiological analysis before market distribution and a very low probability of pathogen's growth during product shelf-life can be predicted by the growth/no growth boundaries produced by the Sym'previus model (Leporq et al., 2005), pathogen cells can survive adapting to stressful conditions. Both Probiotic and Standard products (final pH about 4.0; aw = 0.98) dressed with oil and packaged in modified atmosphere were inoculated with pathogens at a level of about Log 3 CFU/g and stored at 4°C for 45 days. Microbiological (LAB, yeasts, total enterobacteriaceae and aerobic counts) and physicochemical (pH, aw, temperature) analysis were performed in Probiotic and Standard products. The pathogen Sal. enterica subsp. enterica was completely inactivated after 14 and 28 days in Probiotic and Standard products, respectively. E. coli resulted highly inhibited in the Probiotic food reaching the detection limit (DL) of the plate count method (Log 1 CFU/g) at Day 4 while in the Standard products it survived until the end of experiment. The pathogen L. monocytogenes decreased (P<0.05) in the Probiotic product at Day 1 reaching values below the DL after 14 days, while 21 days were needed in the Standard product. However, the presence of the pathogen was evidenced after enrichment steps in both samples until the end of the experimental period. In order to study the behaviour (survival/inactivation) of the three pathogens in the RTE products, the GInaFit program (Geeraerd et al., 2005) was used to fit the available models to the experimental data from the plate count method. Data indicated that pathogens decreased in the Probiotic product in two descent phases, without shoulder and/or tailing, and this behaviour may be related to the exposure of the bacterial cells to the environmental conditions occurring in the presence of the probiotic strain. The biphasic linear model described the survival behaviour of Sal. enterica subsp. enterica and E. coli in the Probiotic product, while the sigmoidal-like model was used to fit the survivor curves of the two pathogens in the Standard product. L. monocytogenes showed a non linear inactivation behaviour and the Weibull and double Weibull models best fitted experimental data from the Standard and the Probiotic product, respectively. Therefore, the different processing conditions used to prepare the Standard and the Probiotic product and in particular the presence of the probiotic strain influenced pathogens' behaviour since the intrinsic and extrinsic factors measured in this study remained stable during the experiment. In fact, the probiotic strain represented always more than the 93% of LAB (about Log 7 CFU/g of product) during the entire experimental period. The L. paracasei LMGP22043 combines the efficacy of a protective culture which can control the development of pathogens during storage with probiotic benefits leading to a final product with functional appeal.