Optimization of the Treatment of Beer Lees for Their Use in Sustainable Biomass Production of Lactic Acid Bacteria

Abstract

Beer lees (BL), a by-product of beer production, consist mainly of dead yeast cells with potential nutritional value. On the other hand, yeast extract (YE), obtained through the lysis of yeast cells, is commonly used as a nutrient-rich supplement for the growth of fastidious microorganisms such as lactic acid bacteria (LAB). However, YE is a high-cost ingredient. Therefore, the aim of this study was to optimize the use of BL as a low-cost alternative source of YE through different lysis treatments, evaluating its suitability to support the growth of UNQLpc 10 and UNQLp 11 strains in a whey permeate (WP)-based medium. Growth kinetics and cell viability were compared with those obtained in MRS broth. The best results were observed with sonicated BL, up to 10 logarithmic units, which supported LAB growth comparable to MRS. Although autolyzed BL promoted lower bacterial growth than sonicated BL, it showed greater cell disruption and higher levels of nitrogen, proteins, and amino acids (5.32%, 26.0%, and 277 nM, respectively). Additionally, autolyzed BL exhibited lower concentrations of reducing sugars and a higher presence of Maillard reaction products, as indicated by colorimetric analysis. These changes, which may be related to the formation of Maillard reaction products during the autolysis process, could have negatively affected the nutritional quality of the extract and, thus, reduced its effectiveness as a bacterial growth promoter.

1. Introduction

Lactic acid bacteria (LAB) have been used in the production of fermented foods for centuries. They can increase the shelf life of food and improve flavor. Among the most versatile and promising LAB species, Lactiplantibacillus plantarum and Lacticaseibacillus paracasei play a critical role in food and agricultural applications and have been associated with numerous health benefits [1,2].

The upscaling of biomass production of LAB as starter cultures for the food industry requires the development of economically viable and environmentally sustainable processes. Large-scale biomass production of selected bacterial strains demands the design of a cost-effective growth medium. Several natural matrices, such as food industry by-products, can be used for this purpose [3,4]. However, supplementation is essential as most food industry by-products are nutrient-poor and require enrichment to achieve significant LAB biomass yields. Commonly used supplements include protein-rich nutritional bases such as yeast extract (YE) and peptone, which are particularly expensive, accounting for nearly 30% of the total production cost [5,6]. Bacteriological YE is industrially obtained from autolyzed yeast cells. Cell lysis can be achieved through several methods, including mechanical disruption, enzymatic hydrolysis, organic solvent treatment, and autolysis. In the latter, enzymes break down the yeast proteins and nucleic acids. Recently, ultrasonication has also been proposed as a means to produce YE. These methods diversify the final product to meet different needs and applications [6,7,8].

As a less expensive alternative to YE, wine lees (spent yeast decanted after alcoholic fermentation of wine) have been proposed as sustainable nutrient sources for biomass production of LAB [9,10]. In contrast, to the best of our knowledge, few studies have explored the use of beer lees (BL) as a nitrogen source for LAB biomass production.

Beer is the most widely consumed alcoholic beverage worldwide. In Argentina, beer production generates approximately 35 million kg of organic waste per year, with around 20 million kg originating in the Province of Buenos Aires. BL, which consist primarily of dead yeast cells, accumulate as a solid residue at the bottom of the wort fermentation tank. The utilization and valorization of this brewing by-product have a direct positive impact on both the economy and the reduction of environmental pollution. BL not only offer the advantages of low production costs and wide availability but may also contain key nutrients, such as amino acids and vitamins [11,12], essential for microbial growth.

Given this background, the objective of this study was to optimize the use of BL as a nitrogen source and growth factor in the biomass production of two LAB species commonly used for food fermentation: Lactiplantibacillus plantarum and Lacticaseibacillus paracasei. It was hypothesized that sonication would represent a superior method of cell disruption, both in terms of efficiency and cost-effectiveness, compared to autolysis, for enhancing the growth of lactic acid bacteria (LAB). The strains evaluated were Lpb. plantarum UNQLp11 and Lcb. paracasei UNQLpc10, both selected for their potential as a malolactic starter culture for winemaking [13,14]. Different cell lysis methods were tested and compared with commercial YE. Finally, the physicochemical properties and amino acid composition of the YE from BL were analyzed.

2. Materials and Methods

2.1. Strain Information

The Lcb. paracasei strain UNQLpc10 (complete genome GenBank Accession code CP029686.1) was isolated from a Patagonian Pinot noir wine vintage 2014 [15], and the Lpb. plantarum strain UNQLp11 was obtained from a Pinot noir wine vintage 2010, chosen for its technological and enological characteristics [13,14]. Both strains are genetic resources belonging to the Province of Río Negro (Argentina), where they were given the registry identification RGPRN-/-M-/-361-/-Semorile-/-626-/-UNQLpc10-/-2024 and RGPRN-/-M-/-357-/-Semorile-/-626-/-UNQLp11-/-2024, respectively.

2.2. Lysis Treatments of Beer Lees

BL, kindly donated by the brewery Bierlife S.A. (Sarandí, Buenos Aires, Argentina), were kept frozen at −20 °C until use. They were evaluated as pure, autolyzed, and sonicated. BL supernatant was also evaluated.

Pure BL: To obtain this, the total BL were sterilized by autoclave and subsequently freeze-dried.

Autolyzed BL: To obtain this, some of the BL were incubated for 48 h at 50 °C (autolysis) [7]. Autolytic activity was stopped by heating (sterilization by autoclave). Then, the BL were centrifuged at 10,000× g for 10 min, and the supernatant was freeze-dried to obtain a dried YE.

Sonicated BL: To obtain this, BL were sonicated using an ultrasonic cleaner (BioBase, Jinan, China) (180 W of power and 40 KHz of frequency, according to the manufacturer’s instructions) for 2 h at 30 °C. Samples were subsequently sterilized by autoclave at 121 °C for 15 min to inactivate all enzymes and microorganisms, then centrifuged at 10,000× g for 10 min, and the supernatant was dehydrated by freeze-drying [8] to obtain the YE.

BL supernatant: To obtain this, BL without treatment (other than freezing and thawing) were sterilized by autoclave at 121 °C for 15 min and then centrifuged at 10,000× g for 10 min. The supernatant obtained was freeze-dried.

All BL samples were freeze-dried to obtain dry powders using freeze-drying equipment (BioBase/BK-FD10P, Jinan, China) for 24 h (condenser temperature: −55 °C; chamber pressure: 0.06 mbar).

2.3. Disruption Efficiency Determination

Disruption efficiency (DE) was determined using the formula described by Avramia et al. [16] based on the loss of dried biomass during lysis treatment:

$$DE=\left(m0-m1\right)/m0$$

(1)

where m0 = dried biomass of the sample before ultrasonic lysis and m1 = residual dried biomass after lysis treatment and centrifugation. An amount of raw dried biomass equal to that resuspended in the case of each sample was freeze-dried using a freeze-dryer Biobase model BK-FD10P and weighed after drying to obtain the m0 values. The supernatants of the centrifuged lysis treatment samples were dried in a similar manner to obtain the m1 values [8].

2.4. Strain Growth

The two strains were grown in sustainable media based on whey permeate (WP). WP was obtained by drying deproteinized sweet whey containing approximately 85% w/w lactose, 6% w/w ashes, and 3% w/w proteins, kindly supplied by Arla Foods Ingredients S.A. (Buenos Aires, Argentina). Table S1 shows nutritional information of WP. WP was used at 5% w/v in distilled water and supplemented according to Cerdeira et al. [17] and YE was replaced by BL lysed with the different treatments mentioned above (

Table 1. A composition of alternative media based on whey permeate (WP).

Medium Composition	Control	Commercial YE	Sonicated BL	Autolyzed BL	BL Supernatant	Pure BL
WP (%w/v)	5	5	5	5	5	5
NH4Citrate (g/L)	2	2	2	2	2	2
K2HPO4 (g/L)	2	2	2	2	2	2
Tween 80 (mL/L)	1	1	1	1	1	1
MgSO4 (g/L)	0.1	0.1	0.1	0.1	0.1	0.1
MnSO4 (g/L)	0.05	0.05	0.05	0.05	0.05	0.05
YE (%w/v)	-	1	1	1	1	1

). Commercial YE (Britania, Buenos Aires, Argentina) was also used as a supplement of WP (Table 1) and MRS broth was used as a positive control [18]. Cultures were incubated at 28 °C under aerobic conditions. Cell viability was determined by bacterial colony counts on MRS agar plates, incubated at 28 °C for 48 h. The data obtained were expressed as CFU/mL.

Growth kinetics were measured by O.D. at 600 nm during incubation for 24 h in a Shimadzu spectrophotometer UV-1603 (Shimadzu Co., Kyoto, Japan). The Boltzmann (sigmoidal) function [19] was used to model growth kinetics:

OD (t) = OD_max/(1 + exp((V50 − t)/Slope))

(2)

where OD (t) denotes biomass, which changes in time t. OD_max corresponds to the top asymptote, slope is the rate of change of OD over time, and V50 is the time where the slope is the maximum. All parameters were determined by GraphPad 8.0 (San Diego, CA, USA).

2.5. Determination of Total and Soluble Proteins and Total Amino Acids

The concentrations of nitrogen, total proteins, trichloroacetic (TCA)-soluble peptides, and total amino acids were measured in the YE obtained from autolyzed and sonicated BL. The total nitrogen content was determined by the Kjeldahl method (AOAC, 1990) using a Kjeltec^MC 8100 distillation module coupled with a DT2508 digestor module and a SR210 scrubber (Foss, Scandinavia, Denmark); the percentage of protein was calculated by multiplying by the factor 6.25.

The protein content was determined by the Lowry method, which was carried out using 2 g of each YE from BL dissolved in 100 mL distilled water and 0.1 mL of the sample, mixed with 1 mL of complex-forming reagent (Na₂CO₃ 20 g/L, CuSO₄·5H₂O 0.1 g/L and sodium potassium tartrate 0.2 g/L). After 10 min at room temperature, 0.1 mL of Folin–Ciocalteu reagent was added, incubated at room temperature for 30 min, and the absorbance at 750 nm was measured [20]. Standard curves were constructed using bovine serum albumin (concentration of 0.1–1%) for proteins and tyrosine for TCA-soluble peptides. The absorbance was measured on a spectrophotometer (UV-VIS 160 A, Shimadzu, Japan).

The amino acids were analyzed by hydrolysis with 6N HCL, under vacuum conditions at 110 °C for 24 h. The hydrolysates were injected into a Biochrom 30 autoanalyzer. The amino acids were separated and quantified by cation exchange and detection by post-column derivatization with ninhydrin [21].

2.6. Measurement of Reducing Sugars

Total reducing sugars were measured using the Fehling–Causse–Bonnans method, which is based on the reduction of the cupric ion in alkaline and hot media by the reducing sugars present in the bagasse [22].

2.7. Color Development by the Maillard Reaction

The Maillard reaction is a complex process resulting in color changes, which results in products with known inhibition of microbial growth. The Maillard reaction in the different YE was measured by colorimetric methods using a computer visualization system [23]. The surface color was measured with a colorimeter (Konica Minolta CR-400, Tokyo, Japan; D65 illuminant, 8 mm aperture, 2° standard observer). Color was expressed as L* (brightness), a* (+a, redness; −a, greenness), and b* (+b, yellowness; −b, blueness). The whiteness (W*) was calculated by the following formula [24]:

W* = 100 – (100 − L*)2 + a* 2 + b* 2

(3)

A total of eight color values were taken from the power surface. Color differences between BL samples and the commercial YE as control were calculated as follows:

ΔE* ab = (Δ L* 2 + Δ a* 2 + Δ b* 2)^1/2

(4)

2.8. Reproducibility of Results and Statistical Analysis

All experiments were carried out using three independent bacterial cultures. The relative differences were reproducible, independently of the culture used. Analysis of variance (ANOVA) was carried out using the statistical program STATISTIX 8 Software (Analytical Software, Tallahassee, FL, USA). Means were compared by Tukey’s or Dunnett’s test for multiple comparisons, and the difference was considered significant when p < 0.05.

3. Results and Discussion

3.1. Growth of UNQLpc10 and UNQLp11

As a first screening to optimize the treatment of BL for use as a nitrogen source, the growth of the two LAB selected (UNQLpc10 and UNQLp11) was analyzed in a sustainable medium formulated with whey permeate, salts, and Tween 80 (WP), and supplemented with 1% w/v of YE from BL treated with different lysis treatments or pure BL. The growth of UNQLpc 10 and UNQLp 11 in the BL media was compared with the growth in MRS broth and in WP supplemented with commercial YE. The growth kinetics of UNQLpc 10 and UNQLp 11 for 24 h are shown in Figure 1A,B, respectively. The number of viable cells expressed in CFU/mL after 48 h of incubation, when all cultures reached the stationary phase, for UNQLpc 10 and UNQLp 11 is shown in Figure 1C,D, respectively, and

Table 2. Parameters of growth kinetics from Figure 1A,B obtained by Equation (2) and cell viability after 48 h (stationary state) from Figure 1C,D.

Parameter	MRS	Commercial YE	Sonicated BL	Autolyzed BL	Supernatant BL
Lacticaseibacillus paracasei UNQLpc10
Maximum OD	1.96 ± 0.03 a	1.71 ± 0.06 b	1.45 ± 0.04 c	0.96 ± 0.08 d	0.82 ± 0.03 d
Slope (h−1)	1.46 ± 0.14 a	3.06 ± 0.50 b,c	2.27 ± 0.31 a,b	3.28 ± 1.04 b,c	4.26 ± 0.57 c
V50 (h)	9.96 ± 0.17 a	10.44 ± 0.50 a,b	10.34 ± 0.35 a,b	12.35 ± 0.92 c	11.51 ± 0.46 c,b
R2	0.996	0.988	0.991	0.960	0.994
Viability at 48 h (CFU/mL)	10.08 ± 0.12 a	10.10 ± 0.23 a	10.15 ± 0.21 a	9.32 ± 0.14 b	8.76 ± 0.09 c
Lactiplantibacillus plantarum UNQLp11
Maximum OD	1.97 ± 0.03 a	1.67 ± 0.13 b	1.45 ± 0.03 c	0.99 ± 0.08 d	1.07 ± 0.04 d
Slope (h−1)	1.23 ± 0.15 a	2.47 ± 0.09 b,c	2.25 ± 0.21 a,b	3.90 ± 0.92 d	3.49 ± 0.34 c,d
V50 (h)	10.71 ± 0.18 a	11.68 ± 0.09 a	11.46 ± 0.21 a	13.33 ± 0.78 b	14.43 ± 0.33 c
R2	0.993	0.999	0.996	0.981	0.996
Viability at 48 h (Log CFU/mL)	9.70 ± 0.25 a	10.03 ± 0.12 a	10.00 ± 0.17 a	9.14 ± 0.16 b	9.04 ± 0.13 b

Different letters denote statistically significant differences.

. Additionally, Table 2 shows the parameters of growth kinetics from Figure 1A,B obtained by Equation (2).

The growth in the control medium (WP supplemented only with salts and tween) was 1.25 log less than that in MRS for both strains, whereas incubation in WP without supplementation showed no growth. The supplementation with commercial YE was required to achieve viable LAB cell levels comparable to those obtained with MRS, without significant differences (Figure 1C,D). These results have been previously reported for UNQLp 11 [17]. However, the growth kinetics differed between MRS and YE-supplemented WP (Table 2) as the latter is a poorer medium than MRS, given that MRS contains a high concentration of glucose and multiple nitrogen sources (peptones and YE), and the growth in MRS is faster and reach stationary state at lower times than YE-supplemented WP (Table 2, Figure 1A,B).

In the case of WP supplemented with BL (lysed through different treatments), several points should be highlighted. First, the addition of pure BL showed no increase in the O.D. (Figure 1A,B); however, surprisingly, the cell viability decreased drastically for both UNQLpc 10 and UNQLp 11 (Figure 1C,D). Both strains decreased 3 log after 48 h of incubation and continued decreasing through time (8 log after 7 days), indicating the presence of an inhibitory component in the pure BL. Beer is known to contain hop acids, which inhibit the growth and survival of probiotic LAB [25]. Although the specific compounds responsible for this inhibitory effect were not quantified in the present study, previous research has indicated that hop-derived acids can strongly inhibit the growth of LAB, and beer lees are known to retain significant amounts of these compounds due to their affinity for yeast cell walls [22]. The lysis treatments, followed by centrifugation, likely helped remove or reduce these inhibitory components. Further studies quantifying hop acids or testing their inhibitory activity directly would be valuable to confirm their specific role in this effect. Unlike pure BL, white wine lees can be directly added to the medium to stimulate LAB growth [26]. Although wine and beer lees may have similar protein nutritional value, their use as a nitrogen source in bacteriology requires different approaches.

Regarding the different methods used for yeast cell disruption, the addition of the sonicated BL supernatant to WP increased the biomass level of both strains, without significant differences in the CFU/mL and growth kinetics, compared to the supplementation of WP with commercial YE (Table 2, Figure 1). The addition of the supernatant (without lysis treatment) showed no significant differences with WP supplemented only with salts and Tween, indicating that freezing and sterilization by autoclave are not enough to induce disruption of yeast cells (Figure 1C,D).

The supernatant from previously autolyzed BL showed worse performance with lower biomass levels than those obtained with sonicated BL, as we can observe in Table 2. The supplementation with autolyzed BL was different for UNQLpc 10 and UNQLp 11 (Figure 1C,D, respectively), UNQLp 11 did not show a difference between autolyzed and supernatant BL (Table 2). This differential response may be attributed to physiological and genomic differences between the strains. Lpb. plantarum and Lcb. paracasei are known for their high metabolic versatility and stress resistance. However, tolerance to various stressors or potential inhibitory compounds—possibly present in the different yeast extract preparations—may be strain-dependent [27,28]. These results show that WP supplemented with sonicated and autolyzed BL had a better performance in the production of cell biomass. Thus, we continued the studies of the composition of sonicated and autolyzed BL (BL-supplemented YE with better results).

3.2. Composition of YE from BL Treated with Different Lysis Treatments

Since different treatments can affect the composition of YE, we then compared several parameters between sonicated BL and autolyzed BL.

The composition of the YE obtained depends on the yeast starting material, which must contain high amounts of specific nutrients. YE is most commonly used in the food industry and appears to be suitable for the normal growth of LAB. This nitrogen source is known to contain a wide range of amino acids and peptides that can meet the requirements of most LAB strains. In addition, YE is not only a source of nitrogen but also a source of carbon, minerals, and vitamins [29].

Thus, we evaluated the amount of nitrogen and amino acid composition in each YE obtained. We also assessed the incidence of the cell disruption treatments on sonicated and autolyzed BL samples. A method to determine lysis efficiency is to quantify the release of intracellular compounds. The higher the release of intracellular compounds, the higher the lysis efficiency.

Table 3 shows that the YE obtained from autolyzed BL has a higher protein and amino acid content than that obtained from sonicated samples, measured by different methods. The lower protein content is consistent with the lower efficiency of the cell disruption, but autolyzed samples have a considerably higher cost and longer time for obtention than sonicated ones: 24 h at 55 °C vs. 2 h at 30 °C for autolyzed vs. sonicated samples, respectively. The protein content estimated by the Kjeldahl method was higher than that obtained using the Lowry assay. This difference is expected as the Kjeldahl method measures total nitrogen, including nitrogen from non-protein sources such as nucleic acids, free amino acids, and other nitrogenous compounds. In contrast, the Lowry method detects primarily polypeptides and proteins, particularly those containing aromatic amino acids such as tyrosine and tryptophan. Therefore, the discrepancy reflects the broader nitrogen profile captured by Kjeldahl analysis compared to the more specific protein focus of the Lowry method. On the other hand, the percentage of peptides was much higher in autolyzed samples, indicating the activity of proteases during this process.

Only the content of total reducing sugars was higher in the sonicated YE samples than that in the autolyzed YE samples, indicating that, after cell disruption, sugars are metabolized during the autolysis process.

Some authors have reported that autolysis has some disadvantages, such as showing low nutrient retention and being more destructive of antioxidant substances, i.e., amino acids, B vitamins, polyphenols, and glutathione [30].

Table 3. Total percentages of nitrogen, proteins, sugars, and amino acids.

YE Treatment	Methods	Autolyzed	Sonicated	% Change vs. Autolyzed BL	References
Nitrogen (%)	Kjeldahl	5.32 ± 0.02	4.08 ± 0.02	↓ 23.31%	-
Total proteins (%)	N × 6.25	33.28 ± 0.13	25.49 ± 0.27	↓ 23.41%	[31]
Total proteins (%)	Lowry	26 ± 1	14.24 ± 1	↓ 46%	[32]
TCA-soluble peptides (%)	Lowry	16 ± 1	7 ± 1	↓ 56%	[32]
Total amino acids (mM)	IEC	277 ± 5	189 ± 6	↓ 31.83%	[21]
Sugars (%)	Fehling	5.88 ± 0.12	10.51 ± 1.19	↑ 78.74%	[33]
Disruption efficiency Time and temperature	-	0.65 ± 0.020 48 h, 55 °C	0.81 ± 0.02 2 h, 30 °C	↓ 24.62% -	[16] -

% = g/100 g of dried sample; TCA: trichloro acetic acid; IEC: ion exchange chromatography.

Table 3. Total percentages of nitrogen, proteins, sugars, and amino acids.

YE Treatment	Methods	Autolyzed	Sonicated	% Change vs. Autolyzed BL	References
Nitrogen (%)	Kjeldahl	5.32 ± 0.02	4.08 ± 0.02	↓ 23.31%	-
Total proteins (%)	N × 6.25	33.28 ± 0.13	25.49 ± 0.27	↓ 23.41%	[31]
Total proteins (%)	Lowry	26 ± 1	14.24 ± 1	↓ 46%	[32]
TCA-soluble peptides (%)	Lowry	16 ± 1	7 ± 1	↓ 56%	[32]
Total amino acids (mM)	IEC	277 ± 5	189 ± 6	↓ 31.83%	[21]
Sugars (%)	Fehling	5.88 ± 0.12	10.51 ± 1.19	↑ 78.74%	[33]
Disruption efficiency Time and temperature	-	0.65 ± 0.020 48 h, 55 °C	0.81 ± 0.02 2 h, 30 °C	↓ 24.62% -	[16] -

% = g/100 g of dried sample; TCA: trichloro acetic acid; IEC: ion exchange chromatography.

The total amino acid percentage in both the autolyzed and sonicated YE samples after acid hydrolysis was determined next. Figure 2 shows the percentages of each amino acid. In general, autolyzed YE samples contained higher percentages of most amino acids, except for glutamate, proline, glycine, and alanine. Proline has been shown to accumulate in vacuoles of yeast cells under stress conditions [34], suggesting that sonication can disrupt these cytoplasmic organelles and leak their content. It should be noted that hydrolysis with 6M HCl for amino acid quantification has some limitations. Acid hydrolysis results in significant losses of glutamine, asparagine, and tryptophan [35]. Other amino acids are destroyed to a lesser extent, as evidenced by the decrease in yields with increasing hydrolysis time. To the best of our knowledge, there is no general consensus on the rate of acid destruction for each amino acid.

Although autolyzed YE contained higher concentrations of amino acids such as isoleucine, leucine, valine, tyrosine, methionine, and phenylalanine—compounds known to support LAB growth [30]—this did not translate into improved biomass production, as shown in Figure 1. This lack of correlation could be attributed to several factors. First, the bioavailability of these amino acids may be reduced by their association with Maillard reaction products (MRPs), which were more abundant in the autolyzed samples, as will be described later (Table 3). MRPs can form complexes with amino acids, reducing their uptake by bacteria [36]. Second, autolysis conditions, particularly prolonged heating, may lead to degradation of heat-sensitive nutrients such as vitamins and peptides that are essential for LAB metabolism [29]. Third, the presence of inhibitory compounds formed during autolysis, such as melanoidins or oxidized amino acid derivatives, could have a negative impact on cell growth despite the apparent nutrient richness [37,38,39]. These observations suggest that nutrient quality and balance, rather than concentration alone, are key factors influencing LAB growth.

Sonication is a highly efficient lysis method [8], and it depends on several parameters previously studied in different works. According to Dimitriu et al. [40], protein release is influenced by the equipment used (by the input acoustic power). Other studies have pointed out that cell suspension concentration is a significant parameter in ultrasonic lysis [8]. In this last study, the authors reported that sample concentration had a notable influence on the disruption efficiency in terms of protein release from lysed yeast cells: the lower the concentration tested, the higher the protein release. In our study, processing times were short (2 h) and the ultrasonic lysis temperature was 30 °C, contributing to the initiation of cell disruption while avoiding the degradation of the valuable biocompounds released [41]. More studies should be performed to improve the efficiency of cell disruption under sonication conditions and its effects on the nitrogen content of YE, as well as the influence on LAB growth. While sonication demonstrated good disruption efficiency and better LAB growth support compared to autolysis, its implementation at industrial scale presents challenges. Ultrasonic equipment suitable for large-scale processing must be able to handle high volumes under controlled temperature conditions to prevent thermal degradation of nutrients. Additionally, sonication processes are typically energy-intensive, particularly in continuous mode. Current advances in flow-through ultrasonic reactors and energy-efficient transducers offer promising avenues for scale-up, but further optimization is required to balance operational costs and lysis efficiency [8,41].

3.3. Maillard Reaction

Another important characteristic of sonicated and autolyzed YE samples is the color of solutions and dried powders, indicating the possible Maillard reaction. Table 3 displays the different parameters of color analysis evaluated, showing a significant decrease in L* and an increase in a* in autolyzed samples. These results have been previously correlated with the Maillard reaction [42].

The ΔE was calculated to better explain the color differences between commercial YE and autolyzed and sonicated BL samples, with the former having the greatest difference with the commercial sample (

Table 4. Color expression in L* (brightness), a* (+a, redness; −a, greenness), and b* (+b, yellowness; −b, blueness) of autolyzed and sonicated YE samples compared with commercial YE.

Color Parameter	Commercial YE	Autolyzed YE	Sonicated YE
L*	80.1 ± 0.7 a	69.1 ± 3.8 c	76.7 ± 2.6 b
a*	−2.37 ± 0.03 a	6.67 ± 1.0 c	1.59 ± 0.8 b
b*	30.42 ± 0.13 a	29.94 ± 1.00 a	28.18 ± 0.81 a
W	63.51 ± 0.28 a	56.45 ± 0.21 b	63.39 ± 0.90 a
ΔE	-	14.38 ± 0.37 a	5.65 ± 0.41 b

Different letters denote statistically significant differences.

).

ΔE was higher in the YE samples from autolyzed BL, which may be due to several causes. The deepening of the color observed in autolyzed samples may be attributed not only to the degradation and oxidation of protein structures but also to the catalytic role of transition metal ions in Maillard reactions. Metal ions such as Fe²⁺ and Cu²⁺ have been shown to accelerate browning by promoting oxidative steps and melanoidin formation [43,44,45].

Maillard reaction products (MRPs) have received considerable attention over the years as potential functional food ingredients that can produce antimicrobial activity through the natural cooking or thermal processing of foods. The antimicrobial activity of MRPs has been attributed to the presence of high molecular weight melanoidins, mostly in sugar-amino acid models, as well as in more complex food systems. MRPs fall into the category of natural biological antimicrobial agents. According to Mu et al. [36], the antimicrobial activity of MRPs appears to depend on the microbial strain and internal/external factors that influence the physicochemical reactions occurring between the microbe and the MRP, although the mechanisms used by MRPs are not entirely clear. Hence, higher MRPs in autolyzed YE could affect the efficiency of this YE on the growth of some lactobacillus strains, as shown in Figure 1. Although these results suggest a negative impact of MRPs on LAB growth, this relationship has not been directly tested in the present study. Future experiments should aim to isolate MRPs from autolyzed yeast extracts and evaluate their antimicrobial activity through direct supplementation assays or antioxidant capacity measurements. These approaches will help elucidate the specific role of MRPs in limiting bacterial biomass production.

4. Conclusions

The use of BL as a rich source of nutrients has been extensively reported. However, their application as a nitrogen source for bacterial growth has been scarcely studied. In the present work, we found that the whole BL had a bactericidal effect on the two LAB strains evaluated, possibly due to the presence of the hop acids added in the beer production process, which have affinity for the yeast wall. The lysis treatment of BL (using different methods) and later centrifugation were enough to reverse this effect. Regarding the treatment, autolysis showed better cell disruption and higher nitrogen content (proteins, amino acids, peptides, and others) but was less efficient in producing LAB biomass. These findings suggest that, in addition to nitrogen, sugars and other sensitive nutrients such as vitamins and polyphenols may play an important role in LAB growth promotion. It is possible that some of these nutrients are degraded during the autolysis process and that the increase in Maillard reaction products contributes to a reduced effectiveness of autolyzed YE as a growth promoter. Finally, the sonication of BL appears to be a sustainable source of nitrogen and growth factor for LAB, avoiding MRPs and eliminating the inhibitory effect of hop acids. Moreover, the valorization of BL contributes to the reduction of organic waste from the brewing industry and may help lower the environmental impact and carbon footprint associated with biomass production for bacterial cultures.