Kinetic pattern and microbial population dynamic characterization of Escherichia coli and Salmonella enteritidis in Frankfurter sausage: An experimental and modeling study

Concerns over the safety of nitrite and its role in the formation of nitrosamines have aroused considerable interest in developing alternative antimicrobial agents to replace nitrite. Plant‐derived antimicrobial agents could serve against food spoilage and pathogens. For the first time in this study, three semiempirical kinetic models, that is, exponential, homographic, and n ‐order kinetic models were established to describe the microbiological population pattern and dynamic behavior. The Escherichia coli and Salmonella enteritidis population dynamics were investigated in Frankfurter sausage as a complex food system, contained four different antimicrobial redcurrant extract (RE) concentrations (0, 0.1, 0.2, and 0.4% wt/wt) at two temperatures (4 and 25°C) during 17 days storage. Furthermore, the results were compared with a neuro‐fuzzy system. Although, all three kinetic models showed a reasonable prediction accuracy, homographic kinetic model was the best for predicting antibacterial effects of RE against E. coli ( R 2 = 0.94–1, R adj 2 = 0.93–0.99, Q ‐squared = 0.80–0.89, and RMSE = 0.21–0.68) and the n ‐order kinetic model did the best in describing the kinetic pattern of S. enteritidis ( R 2 = 0.97–0.99, R adj 2 = 0.94–0.99, Q ‐squared = 0.85–0.93, RMSE = 0.06–0.74). The increase in the RE concentration and temperature resulted in the increase of kinetic parameters of microbial population decline rate and extent for both bacteria, significantly. Furthermore, all the homographic model parameters (except of equilibrium population, p ∞ ) and all the kinetic model parameters with n order (except of reaction order) the synergistic effect of temperature and concentration was clear. RE at 0.4% concentration induced 0.12 log colony forming unit (CFU)/g E. coli and 0.45 log CFU/g S. enteritidis equilibrium population ( p ∞ ) at 17th day of storage at 25°C. Compared to S. enteritidis , E. coli was more sensitive to RE concentration at both temperatures. Practical applications Most formulations of the meat products contain nitrite as a main ingredient to prohibit the pathogenic bacteria growth. The results of the present study showed that all the three studied kinetic models (exponential, homographic, and n ‐order kinetic models) appropriately modeled the S. enteritidis and E. coli population dynamics. Regarding the statistical parameters, n ‐order and homographic kinetic models were found to be the best models for predicting antibacterial effects of RE against S. enteritidis and E. coli , respectively. Furthermore, E. coli was more sensitive against RE than S. enteritidis in Frankfurter sausage samples at 4 and 25°C during 17 days storage. With the increase in the RE concentration at the constant temperature and with the increase in temperature at the constant extract concentration, the decline rate of E. coli population increased significantly ( p  < .05), while the needed time for 50% reduction decreased; besides, S. enteritidis population dynamics showed the increase in the rate and extent of decline, and a decrease in the equilibrium microbial population at these conditions ( p  < .05). No significant differences were observed between the neuro‐fuzzy system results and semiempirical kinetic models in the prediction of microbiological population dynamic. Nevertheless, the novel semiempirical models presented important biological parameters, which provided a better understanding of the systems biological state than the neuro‐fuzzy system. These results could offer new perspectives in the field of microbiology.