Unravelling microbial interactions in a synthetic broad bean paste microbial community

Area of Science:

• Food Microbiology
• Fermentation Science
• Microbial Ecology

Background:

• The microbial ecology of broad bean paste fermentation is complex and not well understood.
• Identifying key microbial interactions is crucial for optimizing fermentation and flavor development.

Purpose of the Study:

• To construct and analyze a synthetic microbial community for broad bean paste fermentation.
• To elucidate the interaction network among key microorganisms using the generalized Lotka-Volterra (gLV) model.
• To apply the interaction network insights to enhance broad bean paste fermentation.

Main Methods:

• Construction of a synthetic community with six key species: Zygosaccharomyces rouxii, Staphylococcus carnosus, Bacillus subtilis, Bacillus amyloliquefaciens, Tetragenococcus halophilus, and Weissella confusa.
• Application of the generalized Lotka-Volterra (gLV) model to predict microbial abundances and analyze interaction networks.
• Sequential inoculation strategy development based on identified microbial interactions.

Main Results:

• The gLV model accurately predicted species abundances in multi-species communities.
• Microbial interactions were dominated by pairwise interactions, with negative interactions being more prevalent (57%) than positive ones (37%).
• Zygosaccharomyces rouxii inhibited acid production, while Bacillus species promoted lactic acid bacteria growth and accumulation.
• The optimized sequential inoculation strategy significantly enhanced ester and mellow flavors, reduced off-odors, and increased volatile flavor compounds by 9.43 times.

Conclusions:

• The gLV model effectively revealed the interaction network of key microorganisms in broad bean paste.
• Understanding microbial interactions is vital for bioaugmentation strategies.
• The developed synthetic community and inoculation strategy significantly improved the flavor quality of broad bean paste.