Extraction and utilization of brewers’ spent grain proteins

The global population is increasing rapidly, driving a higher demand for food. At the same time, consumer preferences are shifting toward sustainability and plant-based diets, which has fueled innovation in the food sector over the past decade. To meet this growing demand in a sustainable way, industries must reconsider how they classify waste, by seeing it instead as a potential source of valuable by-products. This concept aligns with the United Nations’ 12th Sustainable Development Goal, which promotes responsible consumption and production. One industry with significant potential for improved sustainability is the brewing industry. In the brewing process, barley malt is combined with water and heated. The liquid portion is filtered and used for the later fermentation. The solid residue left behind is known as brewers’ spent grain (BSG) and is the primary by-product, accounting for 85% of the total waste generated, or approximately 20 kg of BSG per 100 L of beer. Currently, BSG is mostly used as cattle feed and landfill, but other applications in bakeries have also been exploited lately. Given its high protein content (15–30% dry weight), BSG could serve as a valuable raw material for further processing and protein extraction.

Many food products require stable dispersions of oil and water, known as emulsions. This includes foods such as salad dressings, beverages, and mayonnaise. Emulsifiers are added to stabilize these emulsions by preventing the separation of oil and aqueous phases. Effective emulsifiers are surface-active agents that protect oil droplets from destabilizing forces. Common emulsifiers include proteins, phospholipids, and synthetic surfactants like Tween 20, due to their amphiphilic properties. Another challenge in emulsions is lipid oxidation, which can be detrimental for the shelf life and leads to worsened sensory properties for food containing unsaturated fatty acids. To mitigate lipid oxidation in foods, synthetic antioxidants like ethylenediaminetetraacetic acid (EDTA) or natural alternatives like rosemary extracts are often added. However, the negative opinion towards synthetic ingredients from consumers is leading to a search for new natural alternatives. Identifying new sources can be time consuming and expensive, but predictive tools have been developed to streamline this process. BSG proteins represent a promising alternative to synthetic ingredients, offering a sustainable, plant-based option.

The project aimed to extract proteins from BSG using different pretreatment strategies (microwave, ultrasound, pulsed electric field, ohmic heating, conventional heating), while minimizing the volume of solvent and heat. Furthermore, the project aimed to screen predicted synthetic emulsifying and antioxidant peptides derived from BSG to understand the structure-functional relationship of peptides and to evaluate the use of predictive tools during screening. Two different bioinformatic tools were used: EmulsiPred, and AnOxPePred. EmulsiPred was used to identify embedded emulsifying peptides in BSG proteins, while AnOxPePred was used for antioxidant peptides. Finally, based on the results of the extractions and the results obtained with the validation of the effect of the synthetic peptides, a more targeted enzymatic hydrolysis was carried out, with the aim of releasing the best performing peptides from BSG.

It was possible to extract proteins from BSG without any pH adjustment, however increasing pH to 9 had significant positive effect on the recovery and protein content. Of the tested pretreatment methods, ultrasound showed the highest protein recovery and protein content, whereas pulsed electric field showed no effect on the recovery compared to the control extraction. Microwave treatment also improved protein recovery compared to no pretreatment. The extracts were tested as emulsifiers in 5% oil-in-water emulsions. Although the emulsifying properties of the extracts needed to be improved, as phase separation was observed within three days of storage, the droplet sizes of the pH 9 extracts were significantly smaller than those obtained with the pH 6 extracts. Especially extracts obtained with microwave treatment, ohmic heating, and conventional heating produced emulsions with significantly smaller droplets compared to all other extracts. This suggests that depending on the utilization different pretreatment strategies could be viable.

EmulsiPred was used to identify surface active peptides embedded in the most abundant proteins of BSG to further investigate the emulsifying properties of BSG peptides. EmulsiPred is a bioinformatics tool that can identify amphiphilic peptides, while considering the secondary structure. Based on the analysis, 48 peptides were synthesized by an external peptide provider and tested in emulsions. The results confirmed that EmulsiPred could identify emulsifying peptides, albeit some of the peptides were not able to form stable emulsions, suggesting that the tool could be improved. Furthermore, the results indicated that the peptide chain length significantly influenced the emulsifying activity, but also that the optimal chain length depended on the predicted secondary structure. For α-helical peptides, chain length above 22 amino acid residues led to the highest stability, while medium chain length (8-14 residues) was better for β-peptides. Some of the tested peptides were even able to form more stable emulsions with smaller droplet sizes than sodium caseinate, which is known to be a very potent protein emulsifier.

Synchrotron radiation circular dichroism investigations revealed that the peptides were able to alter their conformation when located at the interface compared to in solution. Furthermore, some of the peptides also showed promising antioxidant activities when tested in fish-oil-in-water emulsions, even though they had not been predicted to have antioxidant activity. Better antioxidant properties were observed for peptides having a positive zeta potential, which could be related to repulsive interactions with the added iron ions. Peptides exhibiting both emulsifying and antioxidant properties are particularly advantageous, as they reduce the need for multiple additives which is also something consumers want.

A similar predictive approach was applied to identify antioxidant peptides using the AnOxPePred tool. As such 25 potential radical scavenging peptides and 25 potential metal chelating peptides were tested. Peptides with high radical scavenging scores were initially screened using the 1,1-diphenyl-2-picrylhydrazil (DPPH) assay. The results showed an IC₅₀ value for the best peptide of 0.024 mg/mL indicating very promising activity. However, when tested in fish oil-in-water emulsions better antioxidant properties were observed for a peptide with IC₅₀ of 2.9 mg/mL. This could be explained by the fact that the latter peptide had a very hydrophobic N-terminal, which could anchor it at the droplet interface, whereas this was not the case of the other peptide. This also means that results from the DPPH assay need to be carefully considered, as the peptides also need to be surface active or present in the oil to show radical scavenging activity in emulsions.

In contrast, metal-chelating peptides act in the aqueous phase to bind metal ions. The screening of the best scoring metal chelating peptides, identified by the AnOxPePred tool, was done using switchSENSE®, a novel biochip method, that can evaluate affinity kinetics of peptides to Ni²⁺. Out of the 25 synthesized peptides, 20 displayed Ni²⁺ affinity, which could indicate metal chelating activity. Eight of the peptides were tested in fish oil-in-water emulsions and resulted in a reduced lipid oxidation rate compared to an emulsion without peptide addition. Notably, the best performing peptide was a negative control of the AnOxPePred tool, with a low metal chelating score, highlighting that the model also could be refined.

The overall results from the synthetic peptide screening indicate that BSG could be a very promising source for production of emulsifying and antioxidant peptides. However, when targeted enzymatic hydrolysis was performed to release high-performing peptides, no increase in emulsifying or antioxidant activities were observed compared to unhydrolyzed samples. This could be due to the low content of protein in the hydrolysates, as well as the fact that single peptides do not dictate the activity of hydrolysates. A viable option could be to do ultrafiltration and fractionation to purify the peptides and remove cellulose and hemicellulose that might interfere with the emulsifying properties.