Extension of growth/no growth predictive models for the preservation of low-acid pasteurized sauces by incorporating water activity and model validation in sauces.

Combined preservation methods have awakened a growing interest in low-acid pasteurized sauces under ambient storage, aiming to produce more 'natural' foods with enhanced microbial stability. However, limited information and predictive models are available to assess the microbial stability of this kind of products, for which the spoilage is mainly caused by acid-tolerant spore-forming spoilage bacteria (ATSSB). In this study, a set of growth/no growth (G/NG) models developed previously (Sun et al., 2021a) for spores of two ATSSB strains (Bacillus velezensis and Bacillus subtilis) from pasteurized sauces were upgraded to incorporate the effect of water activity (a). The growth from heat-treated spores (80 °C, 10 min) was assessed during three months in Nutrient Broth under 320 combinations of storage temperatures (22 and 30 °C), a (0.95 and 0.965), pH (4.4-5.6), total acetic acid (0.0-0.3 % (w/w)), and total lactic acid (0.00-1.00 % (w/w)). Twelve replicates were tested for each combination. After merging the dataset collected previously at a 0.98, a set of 4-variate (4V) models were developed for each strain under a single temperature after 30, 60 and 90 days separately. Additionally, the developed models were validated by challenge tests in different industrially produced low-acid pasteurized sauces for three months. The results showed that lowering a significantly enhanced the antimicrobial effect of decreased pH and the addition of acetic and lactic acid against the ATSSB spores, while the enhancement was less evident at high acetic acid concentrations. In the challenge tests, the developed models exhibited accurate predictions for the sauces, with the % correct-predicted values ≥90 %. The developed models can be used in microbiological risk assessments or the innovation of preservation strategies for low-acid pasteurized sauces.