Predictive modelling of Salmonella: From cell cycle measurements to e-models

Modeling the Inactivation, Survival, and Growth of Salmonella enterica under Osmotic Stress Considering Inoculum Phase and Serotype

AIMS

This study evaluated the behaviour of the Salmonella enterica serotypes in osmotically-stressful BHI broth (0.940 ≤ a_w ≤ 0.960), assessing inoculum from two stages of the bacterial life cycle (exponential and stationary) and two temperatures (25 and 35 °C).

METHODS AND RESULTS

Four S. enterica serotypes (Typhimurium, Enteritidis, Heidelberg, and Minnesota) were grown in stressful BHI at 25 °C. A mathematical model was proposed for describing the total microbial count as the sum of two subpopulations, inactivating and surviving-then-growing. When submitted to a_w of 0.950 and 0.960, all strains showed a decreased count, followed by a period of unchanged count and then exponential growth (Phoenix Phenomenon). Strains inoculated at a_w = 0.940 and 0.945 showed inactivation kinetics only. Cells cultivated at 25 °C and inoculated from the exponential phase were the most reactive to the osmotic stress, showing a higher initial population reduction and shorter adaptation period. The proposed model described the inactivation data and the Phoenix Phenomenon accurately.

CONCLUSIONS

The results quantified the complex response of S. enterica to the osmotic environment in detail, depending on the inoculum characteristic and serotype evaluated.

SIGNIFICANCE AND IMPACT OF STUDY

Quantifying these differences is truly relevant to food safety and improves risk analysis.

Bioinactivation FE: A free web application for modelling isothermal and dynamic microbial inactivation

Mathematical models developed in predictive microbiology are nowadays an essential tool for food scientists and researchers. However, advanced knowledge of scientific programming and mathematical modelling are often required in order to use them, especially in cases of modelling of dynamic and/or non-linear processes. This may be an obstacle for food scientists without such skills. Scientific software can help making these tools more accessible for scientists lacking a deep mathematical or computing background. Recently, the R package bioinactivation was published, including functions (model fitting and predictions) for modelling microbial inactivation under isothermal or dynamic conditions. It was uploaded to the Comprehensive R Archive Network (CRAN), but users need basic R programming knowledge in order to use it. Therefore, it was accompanied by Bioinactivation SE, a user-friendly web application including selected functions in the software for users without a programming background. In this work, a new web application, Bioinactivation FE, is presented. It is an extension of Bioinactivation SE which includes an interface to every function in the bioinactivation package: model fitting of isothermal and non-isothermal experiments, and generation of survivor curves and prediction intervals. Moreover, it includes several improvements in the user interface based on the users' feedback. The capabilities of the software are demonstrated through two case studies using data published in the scientific literature. In the first case study, the response of Escherichia coli to isothermal and non-isothermal treatments is compared, illustrating the presence of an induced thermal resistance. In the second, the effect of nanoemulsified d-limonene on the thermal resistance of Salmonella Senftenberg is quantified.

Nanoemulsified D-Limonene Reduces the Heat Resistance of Salmonella Senftenberg over 50 Times

Salmonella Senftenberg is a pathogen agent causative of foodborne disease and it is considered the most heat-resistant serovar within this genus. Food industries use heat treatment and chemical antimicrobials in order to eliminate this microorganism in food, but consumers prefer natural antimicrobials as essential oils and their components. This study evaluates the combined effect of thermal treatments and different concentrations of D-limonene nanoemulsion on the inactivation of Salmonella (S.) Senftenberg. The results showed an important effect of the nanoemulsified D-limonene on the heat resistance of S. Senftenberg. The δ_{50 °C} value was reduced by 85%, 96% and 98% when 0.1, 0.5 and 1 mM of nanoemulsified D-limonene was added to the heating medium. The effect was kept along all the heating temperatures researched and the shape of the survival curves did not change with the addition of the antimicrobial. The results obtained in this research could be very useful for food industries for optimizing or improving heat treatments applied to food.

Evaluation of indirect impedance for measuring microbial growth in complex food matrices

The suitability of indirect impedance to accurately measure microbial growth in real food matrices was investigated. A variety of semi-solid and liquid food products were inoculated with Bacillus cereus, Listeria monocytogenes, Staphylococcus aureus, Lactobacillus plantarum, Pseudomonas aeruginosa, Escherichia coli, Salmonella enteriditis, Candida tropicalis or Zygosaccharomyces rouxii and CO2 production was monitored using a conductimetric (Don Whitely R.A.B.I.T.) system. The majority (80%) of food and microbe combinations produced a detectable growth signal. The linearity of conductance responses in selected food products was investigated and a good correlation (R(2) ≥ 0.84) was observed between inoculum levels and times to detection. Specific growth rate estimations from the data were sufficiently accurate for predictive modeling in some cases. This initial evaluation of the suitability of indirect impedance to generate microbial growth data in complex food matrices indicates significant potential for the technology as an alternative to plating methods.