Effect of freeze-dried kombucha culture on microbial composition and assessment of metabolic dynamics during fermentation.

Kombucha is a traditional fermented beverage gaining popularity around the world. So far, few studies have investigated its microbiome using next-generation DNA sequencing, whereas the correlation between the microbial community and metabolites evolution along fermentation is still unclear. In this study, we explore this correlation in a traditionally produced kombucha by evaluating its microbial community and the main metabolites produced. We also investigated the effects of starter cultures processed in three different ways (control, starter culture without liquid suspension (CSC), and a freeze-dried starter culture (FDSC)) to evaluate changes in kombucha composition, such as antioxidant activity and sensory analysis. We identified seven genera of bacteria, including Komagataeibacter, Gluconacetobacter, Gluconobacter, Acetobacter, Liquorilactobacillus, Ligilactobacillus, and Zymomonas, and three genera of yeasts, Dekkera/Brettanomyces, Hanseniaspora, and Saccharomyces. Although there were no statistically significant differences in the acceptance test in sensory analysis, different starter cultures resulted in products showing different microbial and biochemical compositions. FDSC decreased Zymomonas and Acetobacter populations, allowing for Gluconobacter predominance, whereas in the control and CSC kombuchas the first two were the predominant genera. Results suggest that the freeze-drying cultures could be implemented to standardize the process and, despite it changes the microbial community, a lower alcohol content could be obtained.

Modelling the growth of Staphylococcus aureus on cooked broccoli under isothermal conditions.

In this study, we developed predictive models describing the growth of Staphylococcus aureus on cooked broccoli florets. A pool of 3.5 log CFU/g of five S. aureus strains were inoculated on 10 g broccoli portions. The samples were then stored at 10, 20, 30 and 37 °C, and colonies were enumerated at different time intervals. Baranyi and Roberts model was fitted to the data using a Bayesian Adaptive Markov Chain Monte Carlo for estimation of the growth parameters. S. aureus showed low growth at 10 °C on broccoli samples and at 20-37 °C interval, Baranyi and Roberts model fitted well to the experimental data (R2>0.97). Estimated growth parameters were correlated with the possibility of toxin production and indicate the potential presence of these biological hazards on contaminated broccoli after heat treatment. Additionally, linear regression was performed for growth rate as storage temperature function. This secondary model followed a linear tendency with R2=0.997 and was compared with two tertiary models (ComBase Predictor and Pathogen Modeling Program) and literature data, demonstrating similar growth rate values of both. These results can be helpful for food services and managers to establish food safety standards for S. aureus growth on cooked broccoli.