Revitalising Brewers' Spent Grains and Enriching With Biogenic Compounds Through the Fermentation of Fructophilic Lactic Acid Bacteria and Yeasts

Abstract

The large output of spent grains from the brewing industry presents environmental concerns but also offers promising nutritional and functional potential for valorization by researchers and industrial stakeholders. In this perspective, we investigated how non-conventional starters like Fructobacillus fructosus PL22 and Wickerhamomyces anomalus GY1 can drive the fermentation of brewer's spent grain (BSG), a solid by-product of the brewing industry, to enrich its portfolio of bioactive compounds. While sugar reduction was comparable between started- and unstarted-BSG, the effect of the fermentation became evident through the release of key microbial metabolites (lactic and acetic acids and ethanol). Both starters generated the highest number of unique peptides, with only one previously identified as antioxidant peptide found in BSG fermented with F. fructosus. During fermentation, most amino acids and phenolic compounds decreased, while BSG fermented with W. anomalus distinctly enhanced the release of Ala, Cys and GABA, and health-promoting phenolic compounds, such as gallic acid, gallocatechin, quercetin, naringenin, kaempferol, and isorhamnetin. These metabolic changes were associated with the enhanced antifungal and antioxidant properties, which in turn positively reflected on skin protection as shown by the increased proliferation of human keratinocytes, over-expression of the filaggrin (FLG) gene, and wound healing. The power of fermentation to revitalise BSG, giving it a second life chance through the improvement of its nutritional value and further multifunctionality, was demonstrated.

Keywords: antifungal activity; antioxidant activity; fructophilic lactic acid bacteria; human keratinocytes; wound healing; yeasts.

FIGURE 1

Cell density of presumptive lactic acid bacteria and yeasts (A) and kinetics of pH (B) of brewer's spent grain (BSG) singly fermented at 30°C for 72 h with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG). BSG incubated under the same conditions, except for the use of starters, was used as the control (Unstarted‐BSG). Data represent the mean ± standard deviation of two biological replicates (each analysed in technical triplicate). Data points with different superscript letters differ significantly (p < 0.05).

FIGURE 2

Carbohydrates, organic acids and ethanol quantification (mg/g FW) in raw brewer's spent grain (Raw‐BSG) and BSG fermented with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG). Fermentation was carried out for 72 h at 30°C. BSG incubated under the same conditions, except for the use of starters, was used as the control (Unstarted‐BSG). Data represent the mean ± standard deviation of two biological replicates (each analysed in technical triplicate). Bars with different superscript letters differ significantly (p < 0.05).

FIGURE 3

Peptidomic analyses of low molecular weight water soluble extracts (LMW‐WSE) obtained from raw brewer's spent grain (Raw‐BSG) and BSG fermented with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG). Fermentation was carried out for 72 h at 30°C. BSG incubated under the same conditions, except for the use of starters, was used as a control (Unstarted‐BSG). The distribution of identified peptides based on the molecular weight, employing a colour scale that transitions from blue to red to represent the Log abundance of each identified peptide within each sample (A); total number and abundance of different peptides found in each sample (B); upset plot of the intersection of samples, sorted by identified peptides, (dark circles in the matrix indicate sets that are part of the intersection) (C); abundance of bioactive peptides found in each sample (D).

FIGURE 4

In vitro hyphal radial growth inhibition (%) against Penicillium roqueforti DPPMA1, Aspergillus versicolor CBS 117286 and Penicillium carneum CBS112297 (A) and ABTS (mM Trolox eq./g) and DPPH (mmol BHT/g) radical scavenging activity (B) of low molecular weight water soluble extracts (LMW‐WSE) and methanol–water soluble extracts (MWSE) obtained from raw brewer's spent grain (Raw‐BSG) and BSG fermented with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG). Fermentation was carried out for 72 h at 30°C. BSG incubated under the same conditions, except for the use of starters, was used as the control (Unstarted‐BSG). Data represent the mean ± standard deviation of two biological replicates (each analysed in technical triplicate). Bars with different superscript letters differ significantly (p < 0.05). P. roqueforti DPPMA1, A. versicolor CBS 117286 and P. carneum CBS112297 without treatments were used as controls. Bars with different superscript letters differ significantly (p < 0.05).

FIGURE 5

Representative images of the effect of raw brewer's spent grain (Raw‐BSG) and BSG fermented with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG) on the migration of human keratinocyte NCTC 2544 cells. Fermentation was carried out for 72 h at 30°C. BSG incubated under the same conditions, except for the use of starters, was used as the control (Unstarted‐BSG). Sub‐confluent monolayers of NCTC 2544 cells were scratched with a sterile P200 pipette tip and treated (at 37°C for 30 h, under 5% of CO₂), with basal serum‐free medium alone (control); 0.5 mg/mL of Raw‐BSG; 0.5 mg/mL of Unstarted‐BSG; 0.5 mg/mL of PL22‐BSG; or 0.5 mg/mL of GY1‐BSG.

FIGURE 6

Filaggrin (FLG) gene expression in human keratinocyte cell line NCTC 2544 through quantitative reverse transcription‐polymerase chain reaction (qRT‐PCR). Cells were treated for 24 (t24) and 48 h (t48) with samples of raw brewer's spent grain (Raw‐BSG), Unstarted‐BSG, or BSG fermented with Fructobacillus fructosus PL22 (PL22‐BSG) and Wickerhamomyces anomalus GY1 (GY1‐BSG). Fermentation was carried out for 72 h at 30°C. Untreated NCTC 2544 cells were used as control. Data represent the mean ± standard deviation of two biological replicates (each analysed in technical duplicate). Asterisk indicates a significant difference (p < 0.05) with respect to the control. Details about the sample preparation are described in the material and methods section.