Sustainability of food side streams: a case study of fermented blends made with sour whey and sunflower press cake powder using the back-slopping technique

Abstract

The exploitation of by-products is a key factor to increase the sustainability of the agri-food chain and fermentation is a simple and eco-friendly process for achieving safe and suitable food materials. In this study, we investigated the possibility to manage a spontaneous fermentation of blends made with different proportions of two food side streams (bovine acid whey and sunflower press cake powder) through the application of a back-slopping technique of the mixed material incubated at 26°C in static conditions. A full-factorial 2-factor 3-level design of experiment was applied to infer the effect of the percent (w/w) of press cake powder in the mixture (20, 25, and 30%) and the rate of back-slopping inoculum (15, 30, and 45%). The pH value, titratable acidity, content of sugars, organic acids, and phenolic acids, enumeration of lactic acid bacteria, yeasts and molds, bacterial contaminants, presumptive Bacillus cereus, and Escherichia coli were measured for each fermentation step at 0, 24, 48, and 72 h. On the same samples, a metataxonomics analysis, targeted on bacterial 16S rRNA gene and fungal ITS region, was performed by using the Illumina MiSeq platform. Acidification of the blends (on average, starting pH = 5.45 ± 011, final pH = 4.61 ± 0.11; starting acidity =13.68 ± 1.02 °SH/50 mL, final acidity = 28.17 ± 2.92°SH/50 mL) and high LAB counts (on average, 9.39 log CFU/g ± 0.25) were observed at the end of each refreshment. In all fermented mixtures, B. cereus, E. coli, and molds counts were lower than the detection limit (<2 log CFU/g), whereas bacterial contaminants, overall spore-formers, were always present (3.74 log CFU/g ± 0.27). After 72 h, the dropping of pH value was maximum, yielding significant differences compared to previous fermentation steps (p < 0.01); particularly, the lowest pH (4.45 ± 0.06) was achieved in the central points of DoE (25% of press cake powder and 30% of back-slopping rate), representing the most suitable condition. Results from both culture-dependent and -independent techniques were consistent; although Lactococcus lactis, continuously deriving from the acid whey, was the main LAB, Pediococcus pentosaceus appeared and, in some cases, became the dominant species. Finally, a long-term trial (about 1 month), using the best condition previously pointed out, was performed with an extension of the incubation time to 84 h for each refreshment. The increase in acidity forced the natural selection toward acid-tolerant microbial strains confirming the former results. Although preliminary, these findings can be useful for developing innovative operations to manage these two relevant side streams implementing the circularity of food resources.