Development and validation of an extensive growth and growth boundary model for psychrotolerant Lactobacillus spp. in seafood and meat products

Abstract

A new and extensive growth and growth boundary model for psychrotolerant Lactobacillus spp. was developed and validated for processed and unprocessed products of seafood and meat. The new model was developed by refitting and expanding an existing cardinal parameter model for growth and the growth boundary of lactic acid bacteria (LAB) in processed seafood (O. Mejlholm and P. Dalgaard, J. Food Prot. 70. 2485–2497, 2007). Initially, to estimate values for the maximum specific growth rate at the reference temperature of 25°C (μref) and the theoretical minimum temperature that prevents growth of psychrotolerant LAB (Tmin), the existing LAB model was refitted to data from experiments with seafood and meat products reported not to include nitrite or any of the four organic acids evaluated in the present study. Next, dimensionless terms modelling the antimicrobial effect of nitrite, and acetic, benzoic, citric and sorbic acids on growth of Lactobacillus sakei were added to the refitted model, together with minimum inhibitory concentrations determined for the five environmental parameters. The new model including the effect of 12 environmental parameters, as well as their interactive effects, was successfully validated using 229 growth rates (μmax values) for psychrotolerant Lactobacillus spp. in seafood and meat products. Average bias and accuracy factor values of 1.08 and 1.27, respectively, were obtained when observed and predicted μmax values of psychrotolerant Lactobacillus spp. were compared. Thus, on average μmax values were only overestimated by 8%. The performance of the new model was equally good for seafood and meat products, and the importance of including the effect of acetic, benzoic, citric and sorbic acids and to a lesser extent nitrite in order to accurately predict growth of psychrotolerant Lactobacillus spp. was clearly demonstrated. The new model can be used to predict growth of psychrotolerant Lactobacillus spp. in seafood and meat products e.g. prediction of the time to a critical cell concentration of bacteria is considered useful for establishing the shelf life. In addition, the high number of environmental parameters included in the new model makes it flexible and suitable for product development as the effect of substituting one combination of preservatives with another can be predicted. In general, the performance of the new model was unacceptable for other types of LAB including Carnobacterium spp., Leuconostoc spp. and Weissella spp.