Answers are grounded in this paper and cite their sources.
Dry-fractionated protein ingredients from pulses are promising for plant-based foods. However, they are co-enriched in polyunsaturated lipids that are susceptible to oxidation, generating beany off-flavors. Fermentation with lactic acid bacteria (LAB) can reduce the abundance of these off-flavors. Although heterofermentative LAB are known to reduce beany aroma more effectively than homofermentative LAB, the magnitude of this efficiency relative to pulse species and degree of fractionation is unclear. Here, we investigated how substrate and microbe selection influence the aroma profiles of LAB-fermented pulse ingredients. We also implemented a novel model coupling volatile organic compound (VOC) profiles with predicted sensory perception using openly available QSAR predictions. Using a full-factorial design, we varied pulse species (faba bean vs. yellow pea), fraction type (starch-rich vs. protein-rich), and bacterial metabolism (obligate heterofermentative Leuconostoc mesenteroides DSM20343 vs. phenotypically homofermentative Lactiplantibacillus plantarum WCFS1). Multivariate analysis of HS-SPME GC–MS data covering 58 VOCs revealed that most variation was explained by pulse species and fraction type, while LAB selection had a smaller but statistically significant effect. Fermentation with heterofermentative LAB removed typical beany aldehydes more effectively than phenotypically homofermentative LAB. Model-based exploratory prediction indicated that these reductions in aldehyde headspace concentrations likely decrease off-flavor perception. Overall, pulse type and protein enrichment strongly determine aroma profiles of LAB-fermented ingredients, but clear efficiency gains arise from heterofermentative LAB. Thus, for a given pulse ingredient, LAB selection remains critical. Our work can support more holistic design of pulse ingredient fermentations by considering the combined effects of substrate and microbial metabolism on final aroma.
No takes yet. Share an insight, caveat, or question.
Novoa et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: