Publication Summary LWT February 1, 2026
Authors: Hang Xiao, Anders Peter Wätjen, Guillermo-Eduardo Sedó Molina, Egon Bech Hansen, Miguel Tovar, Claus Heiner Bang-Berthelsen
Challenge addressed: Plant-based dairy alternatives (PBDA) are nutritionally inferior to cow's milk, particularly in riboflavin (Vitamin B2) – a nutrient essential for human health that the body cannot synthesize. While manufacturers typically add riboflavin artificially, consumer demand for clean-label products creates a need for strains that can produce it naturally during fermentation. Developing new starter cultures for plant-based fermentation is still slower than it should be. If you want strains that actually perform in the real substrate (here: soy) and deliver a nutritional function (riboflavin production), conventional isolation and screening becomes a bottleneck, especially when the best candidates may be rare and embedded in complex communities.
Major achievement: The authors developed a high-throughput, end-to-end droplet microfluidics workflow using Atrandi’s Onyx and Styx platforms and a novel transparent soy-based medium (TSM) for single-cell encapsulation, roseoflavin adaptation, and fluorescence-activated droplet sorting (FADS). Crucially, their approach bypasses initial microorganism isolation, screening all microorganisms in a sample first and then pulling out the winners. This directly isolated riboflavin-overproducing Lactococcus lactis NFICC2835 from bee gut samples, achieving up to 1.23 mg/L riboflavin in commercial soy milks via 5 adaptation-sorting cycles, with genomic analysis revealing key mutations in the rib operon, purH gene, and FMN riboswitch.
Why it mattered: This work addresses a real gap in functional food development and this is a strong example of how compartment-based, high-throughput workflows can accelerate microbial discovery: you can screen directly from a community, enrich functional producers fast, and then validate performance with genomics. The method bypasses slow conventional isolation, preserves microbial diversity at single-cell resolution, accelerates discovery of strains suited for soy fermentation, and enables natural riboflavin enrichment in PBDA to address vegan/vegetarian deficiencies (e.g., EFSA daily needs: 1.3–2.0 mg). NFICC2835 shows versatile sugar metabolism (e.g., starch, phenolics) for broader plant milks, advancing clean-label functional foods and outperforming bulk screening.
Key quotes: “By showcasing the genetic characteristics of the strains obtained from the HTP [High-Throughput] droplet screening, we not only advance our understanding of riboflavin overproduction but also underscore the potential of our screening strategy over conventional approaches,” the team reports.
Explore how Atrandi supports high-throughput screening workflows here.
Read full publication here: https://doi.org/10.1016/j.lwt.2026.119036