Different model hypotheses are needed to account for qualitative variability in the response of two strains of Salmonella spp. under dynamic conditions

Abstract

In this article, the thermal inactivation of two Salmonella strains (Salmonella Enteritidis CECT4300 and Salmonella Senftenberg CECT4565) was studied under both isothermal and dynamic conditions. We observed large differences between these two strains, with S. Senftenberg being much more resistant than S. Enteritidis. Under isothermal conditions, S. Senftenberg had non-linear survivor curves, whereas the response of S. Enteritidis was log-linear. Therefore, weibullian inactivation models were used to describe the response of S. Senftenberg, with the Mafart model being the more suitable one. For S. Enteritidis, the Bigelow (log-linear) inactivation model was successful at describing the isothermal response. Under dynamic conditions, a combination of the Peleg and Mafart models (secondary model of Mafart; t* of Peleg) fitted to the isothermal data could predict the response of S. Senftenberg to the dynamic treatments tested (heating rates between 0.5 and 10 °C/min). This was not the case for S. Enteritidis, where the model predictions based on isothermal data underestimated the microbial concentrations. Therefore, a dynamic model that considers stress acclimation to one of the dynamic profiles was fitted, using the remaining profiles as validation. In light of this, besides its quantitative impact, variability between strains of bacterial species can also cause qualitative differences in microbial inactivation. This is demonstrated by S. Enteritidis being able to develop stress acclimation where S. Senftenbenberg could not. This has important implications for the development of microbial inactivation models to support process design, as every industrial treatment is dynamic. Consequently, it is crucial to consider different model hypotheses, and how they affect the model predictions both under isothermal and dynamic conditions.

Keywords: Dynamic models; Microbial risk assessment; Pasteurisation; Predictive microbiology; Process design.