• Microbial Succession and Flavor Production in the Fermented Dairy Beverage Kefir

      Walsh, Aaron M.; Crispie, Fiona; Kilcawley, Kieran; O'Sullivan, Orla; O'Sullivan, Maurice G.; Claesson, Marcus J.; Cotter, Paul D.; Science Foundation Ireland; SFI/12/RC/2273; SFI/11/PI/1137; et al. (2018-11-05)
      Kefir is a putatively health-promoting dairy beverage that is produced when a kefir grain, consisting of a consortium of microorganisms, is added to milk to initiate a natural fermentation. Here, a detailed analysis was carried out to determine how the microbial population, gene content, and flavor of three kefirs from distinct geographic locations change over the course of 24-h fermentations. Metagenomic sequencing revealed that Lactobacillus kefiranofaciens was the dominant bacterial species in kefir during early stages of fermentations but that Leuconostoc mesenteroides became more prevalent in later stages. This pattern is consistent with an observation that genes involved in aromatic amino acid biosynthesis were absent from L. kefiranofaciens but were present in L. mesenteroides. Additionally, these shifts in the microbial community structure, and associated pathways, corresponded to changes in the levels of volatile compounds. Specifically, Acetobacter spp. correlated with acetic acid; Lactobacillus spp. correlated with carboxylic acids, esters and ketones; Leuconostoc spp. correlated with acetic acid and 2,3-butanedione; and Saccharomyces spp. correlated with esters. The correlation data suggest a causal relationship between microbial taxa and flavor that is supported by observations that addition of L. kefiranofaciens NCFB 2797 increased the levels of esters and ketones whereas addition of L. mesenteroides 213M0 increased the levels of acetic acid and 2,3-butanedione. Finally, we detected genes associated with probiotic functionalities in the kefir microbiome. Our results illustrate the dynamic nature of kefir fermentations and microbial succession patterns therein and can be applied to optimize the fermentation processes, flavors, and health-related attributes of this and other fermented foods. IMPORTANCE Traditional fermented foods represent relatively low-complexity microbial environments that can be used as model microbial communities to understand how microbes interact in natural environments. Our results illustrate the dynamic nature of kefir fermentations and microbial succession patterns therein. In the process, the link between individual species, and associated pathways, with flavor compounds is revealed and several genes that could be responsible for the purported gut health-associated benefits of consuming kefir are identified. Ultimately, in addition to providing an important fundamental insight into microbial interactions, this information can be applied to optimize the fermentation processes, flavors, and health-related attributes of this and other fermented foods.
    • Strains of the Lactobacillus casei group show diverse abilities for the production of flavor compounds in 2 model systems

      Stefanovic, Ewelina; Thierry, Anne; Maillard, Marie-Bernadette; Bertuzzi, Andrea; Rea, Mary; Fitzgerald, Gerald F; McAuliffe, Olivia; Kilcawley, Kieran; Teagasc Walsh Fellowship Programme (Elsevier, 2017-07-12)
      Cheese flavor development is directly connected to the metabolic activity of microorganisms used during its manufacture, and the selection of metabolically diverse strains represents a potential tool for the production of cheese with novel and distinct flavor characteristics. Strains of Lactobacillus have been proven to promote the development of important cheese flavor compounds. As cheese production and ripening are long-lasting and expensive, model systems have been developed with the purpose of rapidly screening lactic acid bacteria for their flavor potential. The biodiversity of 10 strains of the Lactobacillus casei group was evaluated in 2 model systems and their volatile profiles were determined by gas chromatography-mass spectrometry. In model system 1, which represented a mixture of free AA, inoculated cells did not grow. In total, 66 compounds considered as flavor contributors were successfully identified, most of which were aldehydes, acids, and alcohols produced via AA metabolism by selected strains. Three strains (DPC2071, DPC3990, and DPC4206) had the most diverse metabolic capacities in model system 1. In model system 2, which was based on processed cheese curd, inoculated cells increased in numbers over incubation time. A total of 47 compounds were identified, and they originated not only from proteolysis, but also from glycolytic and lipolytic processes. Tested strains produced ketones, acids, and esters. Although strains produced different abundances of volatiles, diversity was less evident in model system 2, and only one strain (DPC4206) was distinguished from the others. Strains identified as the most dissimilar in both of the model systems could be more useful for cheese flavor diversification.