Improving Brewery Sustainability: Upcycling the Discarded Byproducts Trub, Spent Hops, and Yeast as Livestock Feed Additives

Abstract

Craft breweries generate a complex set of byproducts that exceed 2 million tons annually. Their disposal possesses material handling, financial, and environmental challenges. A mixture of these, namely trub, hops, and yeast, designated THYM^®, was evaluated biochemically and in a feeding trial to enhance its valuation. THYM contained approximately 3% α plus β hop acids, 0.2% xanthohumol (XN), and 35% crude protein. It exhibited antimicrobial activity, with a minimum inhibitory concentration of 137 ± 39 μg/mL for B. subtilis, and antioxidant activity, with 90 ± 13 μmol/g of Trolox equivalents. THYM presented positive results in bovine rumen microbial in vitro fermentations, decreasing methane production and the acetate:propionate ratio at 3 mg/mL. These results led to a nine-week feedlot trial with 45 Black Angus weanling steers on either a corn silage-based diet (CON), CON with monensin (MON) at 200 mg/animal, or CON with 1% THYM (THYM). Data were analyzed by SAS 9.4 with two orthogonal contrasts of CON vs. MON and THYM and MON vs. THYM. While average daily gain (ADG) and dry matter intake (DMI) did not differ among treatments, a tendency was observed for the THYM and MON groups to have a greater gain to feed ratio (ADG:DMI) when compared to CON (p = 0.07). The XN metabolite 8-prenylnarigenin, a potent phytoestrogen, was present in the serum of the THYM group at 4.0 ± 0.9 nM by mass spectrometry. These brewing byproducts, which can be drum-dried, were well tolerated and show potential value as a cattle growth promoter.

1. Introduction

Humans have been making beer for 10 millennia as a low-alcohol, potable beverage with nutritional benefits [1], and its annual global production now exceeds 2 trillion hL [2]. The inclusion of hops is documented as of 736 CE, and the annual hop production now is 180,000 t, 95% of which goes to brewing beer [3]. Brewing is a complex process accompanied by a large generation of several byproduct streams that must be managed and disposed of sustainably to minimize environmental impacts [4]. These include wet brewer’s spent grain, brewer’s spent yeast and spent hops, and protein-rich trub. A large volume of wastewater, 3–10 L per L of beer, is also generated [5]. Without proper handling, these byproducts and waste streams contribute to environmental problems such as water and soil contamination and greenhouse gas production [2].

While breweries increasingly attempt to operate sustainably, and spent grain, 33 million t of which is generated annually, is widely used as a valuable livestock feed [6,7,8], the upcycling of other nutrient-rich byproducts—including spent yeast, spent hop and trub—have received far less attention. Spent yeast has well-established uses as a protein source ingredient and food additive; however, its utilization at smaller breweries is often constrained by limited production volumes and a significant water content. A UK study found that 40% to 60% of small and medium urban craft breweries disposed of their spent yeast via municipal sewers [9]. Spent hops and trub have faced similar challenges in feed applications, as well as having the burden of bitterness associated with high concentrations of polyphenols and iso-α-humulones. In addition, the relatively low sales price of spent yeast, spent hops and trub has hindered the development of effective upcycling strategies. Our work focuses on increasing the valorization of these byproducts, particularly in the context of craft breweries, which accounted for 24.7% of US retail beer sales [10].

The flowchart 1A shows the brewing byproducts spent grain, trub (a protein- and polyphenol-rich precipitate), spent hops, and spent yeast. It is noteworthy that a large portion of the hop metabolites shown in Figure 1B are captured in these downstream byproducts principally because of their hydrophobic, low-water solubility properties, as reflected by their high log p values. As much as 85% of added hops, rich in antimicrobial hydrophobic metabolites, end up in byproducts [11]. In other studies, >75% of the β-acids present in the added hops were found in the trub and not in the beer [12], and in another study, the concentration of β-acids in the spent packed yeast paste was 104 μg/mL, >100-fold more than that found in the beer, which was <0.5 μg/mL [13]. The selective accumulation of hop metabolites in spent hops, trub and spent yeast is especially significant in craft brewing where the usage of hops is 3- to 4-fold higher than in multinational breweries [13]. Not surprisingly, the average humulone content of spent yeast from craft breweries was 12 times that of spent yeast from multinational breweries [13]. The growing realization of the value of hop-rich byproducts has led to suggestions regarding methods for extracting or biorefining specific components [14,15] for use in foods. Along those lines, we focus on discovering the benefits that these hop-derived phytochemicals add to animal feeds.

The combination of trub, spent hops, and spent yeast shown in Figure 1A has been termed THYM^® (trub hops yeast mix) and captures these hop metabolites and preserves much of the value of added hops, the costliest beer ingredient on a per kg basis. The antimicrobial [16,17], antioxidant, and other activities of hop metabolites [17,18,19] suggest that combinations of hop metabolite-rich brewing byproducts should have significant biological activity and represent a new value chain if their efficacy can be demonstrated.

In support of this proposition, α and β hop acids appear to share partially overlapping mechanisms of action with monensin (Rumensin^®) on Gram-positive organisms by interfering with microbial potassium fluxes [20,21]. Hop acid-rich spent yeast has been shown to have Gram-positive antimicrobial activity that reduces the production of ammonia and methane by rumen microbes in vitro [22,23]. Hop cones, pellets, and extracts of hops (Humulus lupulus) have shown positive in vitro effects in bovine rumen microbial fermentations, with increased propionic acid formation and decreased methane formation [24,25,26,27].

Feeding trials in beef cattle supplemented with hops have shown promise, with a numerical increase in average daily gain (ADG) for animals on the highest hop level (a hop β-acids feed content of 80 mg/kg) [28]. However, no effects occurred in two other beef cattle feeding studies [29,30].

Hop β-acids possess in vivo activity in other species. Chicken feed containing ground hop cones providing 23 mg/kg β-acids improved feed conversion ratio [31]. The potassium salt of hop β-acids added to drinking water (125 mg/L) reduced bacterial counts in midgut and cecum for Clostridium perfringens subgroup and lactobacillus group counts in chickens [32]. Encapsulated hop β-acids in feed (30 mg/kg) improved growth performance in chickens challenged with a coccidiosis vaccine [33] and positively altered the instestinal microbiota of chickens challenged with Eimeria sp. [34]. In the case of swine, encapsulated hop β-acids at 20 mg/kg feed increased the ADG and gain to feed (G:F) of weanling pigs over a 35-day period [35].

The amount of brewing byproducts, trub, spent hops and yeast potentially available as growth supplements is large, approximately 7 million t [2]. The global market for hops in 2024 was $9.2 billion USD, and is expected to rise to $12.5 billion by 2031 https://www.cognitivemarketresearch.com/hops-market-report (accessed 14 March 2026). As shown in Figure 1A, a large portion of hop metabolites end up as trub, spent hops, and yeast.

The objective of this study was to demonstrate the value of the craft brewing byproducts trub, spent hops, and spent yeast as potential feed supplements and possible alternatives to ionophore growth promoters [36,37], in part by use of in vitro antimicrobial and antioxidant assays and fermentation studies with bovine rumen microbes. Promising results from these studies led to a growth study using Black Angus steers. We hypothesize that THYM supplementation at 1% DM will improve feed efficiency with efficacy like monensin.

2. Materials and Methods

2.1. Collection, Analysis and Drying of Byproducts (THYM)

Brewing byproducts were obtained at Highland Brewing Co., Asheville, NC, USA. Hot trub [38] was collected from the whirlpool. Spent yeast and hops were collected at the end of fermentation by drainage from chilled tanks [11]. These combined byproducts were collected from a variety of beer styles including porters, stouts, India pale ales and pilsners. Collections were prepared weekly, pooled in 1041 L tote containers, occasionally stirred by hand for 5 min at a time to remix, and kept at 15 °C to 20 °C. The slurry preparations had a water content of >85%, as determined by drying to constant weight for 2–3 h at 105 °C. Slurry samples were sent to Dairy-One Forage Lab, Ithaca, NY, USA and Cumberland Valley Analytical Services, Waynesboro, PA, USA for nutritional analysis.

Hop (Humulus lupulus) metabolite analysis of byproducts was conducted by high-performance liquid chromatography (HPLC) by the extraction of 1 mL of THYM slurry with 9 mL of methanol/phosphoric acid (100/1, v/v) in a polypropylene tube, which was sonicated for 10 min and shaken for 20 min at room temperature. A portion of the liquid was clarified in a microcentrifuge at 13,000× g and analyzed by HPLC on a Shimadzu system (Columbia, MD, USA) with dual wavelength ultraviolet (UV) detection fitted with a C18, Brownlee Pecosphere 3μ, 83 × 4.5 mm column, manufactured by PerkinElmer (Shelton, CT, USA) [23]. The HPLC Solvent A contained water/methanol/phosphoric acid (50/50/0.5, v/v/v), and Solvent B contained methanol. The run began at 70% solvent B for 5 min, then a linear gradient was started to reach 100% B by 10 min, and was held there until 14 min. The flow rate was maintained at 1 mL/min, and UV absorbances were measured at 270 nm (iso-α-acids, α-acids and β-acids) and 335 nm (XN, α-acids and β-acids). The UV detector data was collected and chromatographic peaks manually integrated with LoggerPro 3.16.2 analysis software by Vernier Software and Technology, Beaverton, OR, USA.

The HPLC reference standards included CO₂ hop extracts, containing 61.6% α-acids (humulones) and 15.7% β-acids (lupulones), kindly provided by S. S. Steiner (New York, NY, USA). The dicyclohexylammonium (DCHA) salts of trans-iso-α-acids, DCHA humulinones, and DCHA hulupones were obtained from the American Society of Brewing Chemists, St. Paul, MN, USA. Metabolite concentrations in extracts were determined by area integrations and 5-point standard curves prepared as a series of 3-fold dilutions of standard compound mixtures in methanol. For experiments in Table 1, the separated α- and β-acid fractions, tested for antimicrobial and antioxidant activity, were isolated from the CO₂ hop extract mixture on 100–200 mesh silicic acid (Sigma-Aldrich, St. Louis, MO, USA), with a stepwise gradient of hexane and ethyl acetate. XN was obtained from Dr. Stevens.

Byproduct slurries (>90% moisture) were dried on a Buflovak Double Drum Dryer, Buflovak, Tonawanda, NY, USA [39]. For the extraction of drum-dried byproducts, 2.5 g was stirred for 3 h with 10 mL of 95% ethanol and the suspension centrifuged at 3000× g for 10 min to obtain an extract. The weight of the extracted residue was 0.45 g based upon the weight of the residue left on taking a sample of the extract to constant weight under a stream of nitrogen.

2.2. In Vitro Biological Activity Measurements

2.2.1. Measurement of Antimicrobial Activity with Bacillus subtilis

Antimicrobial activity was determined by one of two ways. The first was by Agar diffusion [23,40] on LB Broth with agar (Lennox, Mannedorf, Switzerland) from Sigma-Aldrich where ethanol extracts or standards in organic solvents were prepared in ethanol in serial 5-fold dilutions, and 20 μL of each dried on sterile 6 mm paper disks (Carolina Biological, Burlington, NC, USA), which were then placed on agar plates seeded with Bacillus subtilis and incubated for 18 h at 39 °C. In the other, microtiter broth dilution assays [41] were used to determine minimum inhibitory concentration (MIC) and were performed in sterile Nunc U bottom 96 well plates (Thermo Fisher Scientific, Waltham, MA, USA) with Luria Broth Base media (Invitrogen, Carlsbad, CA, USA) and the aerobic Gram-positive Bacillus subtilis bacteria [42] (Presque Isle Cultures, Erie, PA, USA). For the assay, concentrated compounds and extracts initially in DMSO or ethanol were diluted 20-fold in water, followed by serial 2-fold dilutions in water. For the assay, 50 μL of diluted sample, 150 μL of media (LB Broth), and 50 ul of a B. subtilis stationary culture (A₆₀₀ 5 to 30 mAU) were added to plates, mixed by shaking for 20 min, and incubated for 14–18 h at 39 °C without shaking. Absorbance at 600 nm was read on a Tecan Infinite M200 Pro plate reader (Tecan Instruments, Mannedorf, Switzerland). Controls for sample absorbance, solvent effects, and uninhibited growth were included. The MIC value was taken as the last well of a sample serial dilution set with growth inhibition of >90%.

2.2.2. Measurement of Antioxidant Activity with ABTS

The assay used 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid (ABTS) (Sigma-Aldrich) in a radical scavenging assay [18], with Trolox (Sigma-Aldrich) or ascorbic acid (Thermo Fisher Scientific) as reference standards. Drum-dried THYM samples were extracted as described above, except that 1.5 g samples were stirred with 10 mL of either 70% ethanol:30% water, 40% methanol:60% water, or water alone. The Trolox equivalent antioxidant capacity (TEAC) of samples was determined from the Trolox standard curve as μmole Trolox/g dry material.

2.2.3. Rumen Fluid Fermentations, Short Chain Fatty Acid (SCFA), and Methane Measurements

In vitro fermentations utilized rumen contents from a fistulated steer that was predominantly fed a forage-based diet housed at the NCSU Metabolism Unit, Raleigh, NC, USA. The steer was maintained according to the guidelines of the Institution of Animal Care and Use Committee (IACUC). The contents were filtered through a double layer of cheese cloth and the rumen fluid was diluted 1:2 with a buffer that contained in 1 L of deionized water, the following: NaHCO₃, 3.675 g; NaH₂PO₄, 5.775 g; KCl, 0.360 g/L; CO(NH₂)₂, 0.300 g; NaCl, 0.282 g; MgSO₄:7H20, 0.036 g and CaCl₂:2H₂O, 0.018 g A total of 30 mL of diluted rumen fluid, 1 g of ground alfalfa/corn (3/2, w/w) and 0, 20, 40, 80 or 100 mg (DM) of THYM as a slurry (DM approximately 15%) was added to 125 mL Wheaton fermentation bottles while they were continuously sparged with CO₂. The bottles were closed with screw caps containing a rubber septum for gas sampling and placed in a 37 °C water bath with occasional stirring for 24 h. For the zero-time controls, an identical set of bottles was prepared but immediately put on ice. Fermentations were performed in triplicate. All samples were tested for pH, and for the production of short chain fatty acids (SCFAs) and methane. SCFA measurements were performed by gas chromatography-flame ionization detection on a 30 m, 0.51 mm ID, 0.1 Supelcowax 10 column (Sigma-Adrich), with 2-ethylbutyric acid (Sigma-Adrich) as an internal standard [43]. Methane produced during incubations was measured by gas chromatography with flame ionization detection as described [44].

2.3. In Vivo Biological Activity of THYM

The study with weanling Black Angus steers was conducted at the North Carolina Department of Agriculture and Consumer Services, Mountain Research Station (MRS), in Waynesville, NC, USA. It utilized protocols approved by the Institutional Animal Care and Use Committee (IACUC) at North Carolina State University, Raleigh, NC, USA (16-212-A).

2.3.1. Experimental Design of Feedlot Study with Weanling Black Angus Steers

Forty-five purebred Black Angus steers were used in the study. The experimental design was dictated by the MRS facility consisting of 15 half-covered dry lot feeding pens and Black Angus steers, with a maximum effective capacity of 45 cattle. The experiment was run as a completely randomized controlled trial for 9 weeks with three groups: control (CON), monensin treatment (MON), and THYM treatment (THYM), with five replications of each treatment. An experimental unit comprised three steers in a 18 × 4 m pen with a 4 m feeding trough and ad libitum water. Black Angus steers with a body weight (BW) = 302 ± 21 kg were sourced from the Upper Piedmont Research Station in Reidsville, NC, USA, and the MRS, Waynesville, NC, USA. Upon arrival, steers were commingled and stocked on tall fescue and orchard grass mixed pastures until the beginning of the adaptation period. Steers were then blocked by origin and body weight, and all 45 were randomly assigned to one of 15 pens, 3 × m² per animal. MRS animals were weighed on a Q Catch chute with Tru-Test™ scales (Tru-Test Datamars, Lamone, Switzerland). Steers were then given ad libitum access to corn silage, mixed grass hay, and water for a 14-day adaptation period prior to the start of the 63-day feeding trial. The trial was conducted between October and December 2018. At the end of the adaptation period, each pen was assigned to one of three dietary treatments. Initial BW was recorded as an average of the empty BW taken on two consecutive days at the beginning of the experimental period. BW was recorded every 21 days thereafter, yielding a total of four BW measurements. For the feedlot study, THYM slurry was transported from Highland Brewing Co. to the MRS in 20 L buckets and stored at 10 °C to 15 °C until mixed into the total mixed rations (TMRs). THYM preparations were used within 1 week. THYM batches were analyzed for nutrients and hop metabolites as described above. Dietary treatments were treated as total mixed ration (TMR) additives, with the base of the control consisting of corn silage, ground corn, and soybean meal, with no additive for the control rations (CON) or with the addition, as a premix, of monensin sodium (Rumensin-90, Elanco, Indianapolis, IN, USA) at a level of 30 mg/kg to achieve 200 mg/day/animal (MON), or THYM at 1% final amount in feed. TMRs were based on corn silage provided by the MRS and were formulated according to the National Research Council (NRC) 1996 USDA requirements for 0.9 kg of ADG with an expected dry matter intake of 2.5% of BW. Diets were prepared daily to provide equal amounts of crude protein and total digestible nutrients across treatments. The feed ingredients—corn silage, hay, ground corn, and soybean meal—in the THYM group were decreased slightly to keep diets isonitrogenous. Ad libitum water was available to all steers throughout the experiment. At approximately 0800 h each day, bunkers were swept clean, refusals were weighed on Ohaus scales (Parsippany, NJ, USA), and a subsample was placed in a 100 °C forced air oven (Shel Lab; Sheldon Manufacturing Inc., Cornelius, OR, USA) for determination of refusal dry matter. Pen feed allotment was adjusted each day based on the previous day’s consumption. Immediately following refusal collection, diets were mixed in a Jaylor 5100 batch mixer (Jaylor, East Garafraxa, ON, CA, USA) for 15 min prior to feed out. Dispensed feed weights were recorded on the Jaylor 5100. All mixing equipment were washed down between diet preparations to prevent MON and THYM contamination.

2.3.2. Blood Collection and XN Blood Metabolite Analysis by LC-MS-MS

Prior to the morning feeding, blood samples were collected via jugular venipuncture on day 63 into 6 mL sterile vacutainer blood collection tubes without additives (Vacutainer; Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Tubes were immediately placed on ice until sample collection was completed. Samples were then transported to the MRS laboratory and rested at room temperature for 30 min, before being centrifuged at 2000× g at 4 °C for 30 min. Serum was collected and samples were stored at −18 °C. Portions of the serum samples were extracted and analyzed by liquid chromatograph tandem mass spectrometry (LC/MS/MS) [45] using ¹³C-labeled as internal standards.

2.4. Statistical Analysis

2.4.1. In Vitro

In vitro data were analyzed by two-tailed Student’s t-test, or by Proc GLM procedure of SAS 9.4 (SAS Institute, Inc., Cary, NC, USA).

2.4.2. In Vivo

The in vivo growth performance and feed intake data of Black Angus steers were analyzed using the MIXED procedure of SAS 9.4, with pen (N-5) as the experimental unit. The model for performance data included treatment as a main effect. Performance data was analyzed using repeated measures to describe overall trends, and analyzed separately by period to account for variations in the nutrient composition of THYM. Results were presented as least square means ± standard error of the mean (SEM). Statistical significance and tendencies were reported as p < 0.05 and p ≥ 0.05 but ≤0.10, respectively. If a treatment effect of p ≤ 0.20 was recorded, two orthogonal contrasts were performed to analyze the effect of CON vs. MON + THYM, and to analyze MON vs. THYM.

3. Results

3.1. Collection and Analysis of THYM Preparations and Drying Experiments

The mixture of the brewing byproducts trub, spent hops, and spent yeast, termed THYM, was found to be rich in hop metabolites, as shown by HPLC–UV analysis comparing the elution of standards (Figure 2A) and of a methanol extract of THYM (Figure 2B). The concentrations on a dry matter (DM) basis of the principal hop acid metabolites in this sample were iso-α-acids, 4.5 mg/g, XN, 0.3 mg/g, α-acids, 14.3 mg/g, and β-acids, 15.8 mg/g. The collection process of THYM, described in Materials and Methods, yielded a slurry with a water content of 88% to 95%. While the THYM slurry could be used as such, it was found that it could be de-watered by drum drying, with a good recovery of heat labile hop acids [39]. In three separate runs, using portions from one preparation of THYM slurry, the moisture content was reduced from 95% to 5 ±1%, with good recovery of hop metabolites: XN, 94 ±10%, α-acids, 82 ± 6%, and β-acids 77% ± 4%. The drum temperatures for the three runs were between 89° C and 96° C, and slurry contact time with the drum was approximately 90 s.

3.2. Characterization of THYM Preparations

3.2.1. In Vitro Antimicrobial Activity

Ethanol extracts of THYM, spotted on disks, showed strong antibiotic activity (Figure 3A) in agar diffusion against the Gram-positive bacteria, Bacillus subtilis, which showed clear decreasing size zones of inhibition at 600 μg, 180 μg, 36 μg, and a slight zone at 7 ug on the disks. The β-acid standard gave clear zones of decreasing size at 30, 6 and 1 ug. Microtiter dilution assays were performed with Bacillus subtilis and the ethanol extract to determine a minimum inhibitory concentration (MIC) of THYM, as shown in Table 1. A value of 137 ± 39 μg/mL was determined, calculated as the dry extracted THYM. The MICs for the standards, α-acids, β-acids, and monensin were 8, 2, and 3 μg/mL, respectively.

Figure 3. Agar diffusion assays of a β-acids standard and an ethanol extract of THYM, performed as described in Materials and Methods. Left (A), a stock solution of purified β-acids in DMSO; 7.6 mg/mL was diluted in 5-fold steps with acetone, and 20 uL of each dried on sterile paper disks. Right (B), an ethanol extract of THYM; 45 mg/mL was diluted in 5-fold steps with ethanol and dried on disks as above. Disks were placed on agar plates seeded with B. subtilis and incubated for 18 h at 39 °C.

Figure 3. Agar diffusion assays of a β-acids standard and an ethanol extract of THYM, performed as described in Materials and Methods. Left (A), a stock solution of purified β-acids in DMSO; 7.6 mg/mL was diluted in 5-fold steps with acetone, and 20 uL of each dried on sterile paper disks. Right (B), an ethanol extract of THYM; 45 mg/mL was diluted in 5-fold steps with ethanol and dried on disks as above. Disks were placed on agar plates seeded with B. subtilis and incubated for 18 h at 39 °C.

Table 1. Antimicrobial and antioxidant activity of THYM and reference compounds.

Sample	Antimicrobial Activity 1,2		Trolox Equivalent Antioxidant Capacity 3,4
	MIC 5 (μg/mL)	SD	μmole-/g	SD
THYM 3	137 c	39	90 d	13
α-Acids	8 b,d	3	1140 a	220
β-Acids	2 a,d	1	263 c	16
Iso-α-acids	>250		1 e	1
Monensin	3 d	3	ND
Vitamin C	ND		494 b	91

^a–e Within a column, means without a common superscript differ, p < 0.05 (two-tailed Student’s t-test). ¹ Broth dilution for antimicrobial assay with B. subtilis was performed three times. ² The MIC values are calculated based upon the amount of dry THYM in the ethanol extract to produce an effect. For the antimicrobial assay, 2.5 g of dry THYM was extracted with 10 mL of 95% ethanol and aliquots tested as described in Materials and Methods. ³ The antioxidant assay utilized quenching of the 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical, performed in triplicate, and the Trolox Equivalent Antioxidant Capacity (TEAC) values are calculated based upon the amount of dry THYM in the ethanol extract. ⁴ For the antioxidant assay, 1.5 g was extracted with 10 mL of 70% ethanol and aliquots tested as described in Materials and Methods. ⁵ MIC, minimum inhibitory concentration; ND, not determined.

Table 1. Antimicrobial and antioxidant activity of THYM and reference compounds.

Sample	Antimicrobial Activity 1,2		Trolox Equivalent Antioxidant Capacity 3,4
	MIC 5 (μg/mL)	SD	μmole-/g	SD
THYM 3	137 c	39	90 d	13
α-Acids	8 b,d	3	1140 a	220
β-Acids	2 a,d	1	263 c	16
Iso-α-acids	>250		1 e	1
Monensin	3 d	3	ND
Vitamin C	ND		494 b	91

^a–e Within a column, means without a common superscript differ, p < 0.05 (two-tailed Student’s t-test). ¹ Broth dilution for antimicrobial assay with B. subtilis was performed three times. ² The MIC values are calculated based upon the amount of dry THYM in the ethanol extract to produce an effect. For the antimicrobial assay, 2.5 g of dry THYM was extracted with 10 mL of 95% ethanol and aliquots tested as described in Materials and Methods. ³ The antioxidant assay utilized quenching of the 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical, performed in triplicate, and the Trolox Equivalent Antioxidant Capacity (TEAC) values are calculated based upon the amount of dry THYM in the ethanol extract. ⁴ For the antioxidant assay, 1.5 g was extracted with 10 mL of 70% ethanol and aliquots tested as described in Materials and Methods. ⁵ MIC, minimum inhibitory concentration; ND, not determined.

3.2.2. In Vitro Antioxidant Activity

A 70% ethanol:30%water THYM extract showed strong antioxidant activity, with a TEAC value of 90 μmol/g of dry extracted THYM by the ABTS radical-quenching assay (Table 1). In other THYM extractions, using either 40% methanol: 60% water or water alone, the TEAC values were 51 and 6 μmol/g, respectively. For comparison, vitamin C gave a value of 494 μmol/g (Table 1).

3.2.3. In Vitro Fermentations with Bovine Rumen Fluid

The effect of THYM on metabolic activity in mixed culture in vitro fermentations with bovine rumen microbes and alfalfa:corn substrates is shown in

Table 2. THYM causes a dose–response reduction in methane production and A/P ratio in fermentations by mixed cultures of rumen microbes ¹.

Item	THYM, mg/mL						p Value
	0.0	0.7	1.3	2.7	3.3	SEM	Linear	Quadratic
Hour 0
Total SCFA, mM	66.1	67.3	66.2	66.9	63.2	1.86	0.94	0.87
pH	6.76 a	6.72 a	6.70 a	6.55 b	6.56 b	0.03	0.01	0.14
Acetate, mol %	64.7	65.3	65.5	65.3	65.3	0.33	0.13	0.35
Propionate, mol %	17.9	17.5	17.3	17.2	17.2	0.17	0.01	0.44
Butyrate, mol %	11.2	11.3	11.1	11.4	11.2	0.12	0.80	0.31
Valerate, mol %	2.5	2.5	2.5	2.4	2.6	0.07	0.94	0.73
A/P	3.6	3.8	3.8	3.8	3.8	0.06	0.05	0.40
Methane, nmol/mL	24	22	20	25	32	3.00	0.97	0.14
Hour 24
Total SCFA, mM	166.6	169.4	178.9	172.7	165.8	6.6	0.51	0.16
pH	5.38 a	5.45 b	5.47 bc	5.55 d	5.51 cd	0.01	0.01	0.97
Acetate, mol %	43.6	43.9	44.3	44.2	42.7	0.52	0.78	0.23
Propionate, mol %	30.2	31	32.1	36.8	36.8	0.76	0.01	0.06
Butyrate, mol %	23.3	22.2	20.7	15.9	17.3	0.83	0.01	0.26
Valerate, mol %	1.4	1.5	1.6	1.7	2.1	0.06	0.01	0.42
A/P	1.43 a	1.40 a	1.43 a	1.20 b	1.17 b	0.04	0.01	0.01
Methane, nmol/mL	430 a	392 ab	341 b	345 b	342 b	21	0.01	0.16

^a–d Within a row, means without common superscripts differ, p < 0.01. (Proc GLM procedure of SAS). ¹ Bovine rumen obtained from a fistulated steer was diluted 1:2 with the buffer salts described in methods and 30 mL of this solution was incubated with 1 g of ground alfalfa:corn (3:2) and THYM (dry matter) in a 125 mL Wheaton fermentation bottle, sparged with CO₂, closed with a septum containing screw caps, and placed in a 37 °C water bath with occasional stirring for 24 h. An identical set of samples was immediately put on ice and kept a 0 °C for the zero-time control. Fermentations were performed in triplicate.

. The fermentations were metabolically productive, with an increase in total SCFA from 66 mM to 166 mM over 24 h, a pH drop from 6.8 to 5.4, and an increase in methane production from 24 to 430 nmol/mL of incubation fluid (Table 2). The dose–response study (Table 2) indicated that adding THYM in concentrations as little as 1.3 mg/mL reduced in vitro methane production, and 2.7 mg/mL altered the SCFA profile. An increase in fermentation pH was seen at higher levels of THYM addition (Table 2).

3.3. In Vivo Feeding Trial

Performance

The feeding trial with Black Angus cross-bred steers placed the animals in three diet groups, CON, MON, and THYM, as noted in Material and Methods. The collections of THYM for the feeding trial were prepared weekly at the Highland Brewing Co. and were analyzed for hop acids and nutrients (

Table 3. Composition of THYM preparations used during the animal study ¹.

Item	Period 1	Period 2	Period 3
	g/100 g DM 2			SEM
Moisture a	89.6	85.9	87.1	0.6
CP	33.8	35.7	35.2	0.7
Calcium	0.9	0.8	0. 9	<0.1
Phosphorus	1.2	1.5	1.5	0.1
Ash	7.6	7.3	7.9	0.1
NDF	23.3	21.3	22.6	0.4
ADF	18.6	14.4	17.0	1.0
	mg/g
iso-α-acids	5.3	4.9	6.9	0.1
α-acids	29.9	21.9	24.3	1.0
β-acids	16.7	14.6	14.3	4.2
XN	1.5	1.0	1.8	2.4
	Units
pH b	5.03	5.63	6.50	0.37

^a The moisture values of periods 1 and 3 differed, p = 0.038. ^b The pH values of periods 1 and 3 differed, p = 0.004 (two-tailed Student’s t-test). ¹ A total of 120 L–150 L of THYM slurry was collected weekly for a total of nine preparations during the study; samples of each were analyzed for nutrients and hop acids as described in the Methods. Means and standard deviations are shown for the three preparations used for each 3-week period. ² DM, dry matter; CP, crude protein; NDF, neutral detergent fiber; ADF, acid detergent fiber; XN, xanthohumol.

). The results show that weekly THYM preparations, averaged for each period, were consistent except for small differences between Periods 1 and 3 regarding moisture content and pH. The TMR compositions for the three treatments are shown in

Table 4. Feed and chemical composition of total mixed rations formulated without a feed additive (control), or with the addition of monensin or THYM.

Item	Treatment
	Control	Monensin	THYM
Ingredient	g/100 g DM 1
Corn silage	66.97	66.90	66.34
Mixed grass hay	7.54	7.53	7.47
Ground corn	10.92	10.90	10.81
Soybean meal	12.48	12.47	12.36
Mineral premix	1.10	1.10	1.09
Limestone	0.99	0.99	0.98
THYM	-	-	0.95
Monensin	-	0.003	-
Formulated composition
DM	32.13	33.10	32.41
CP	13.26	13.15	12.73
Calcium	0.83	0.84	0.80
Phosphorus	0.31	0.32	0.32
Starch	26.99	28.88	26.27

¹ DM, dry matter; CP, crude protein.

.

Animal growth performance data for the three diet treatments were measured every 21 days and are presented in

Table 5. Growth performance and feed intake of steers fed a total mixed ration without a feed additive (control (CON)), or with the addition of monensin (MON) or THYM (THYM).

Item	Treatment			Contrasts
	CON	MON	THYM	SEM	CON vs. MON and THYM	MON vs. THYM
	ADG 1, kg/d				-------- p < -------
Total	1.54	1.65	1.72	0.09	0.22	0.6
Period 1	1.34	1.43	1.53	0.12	0.34	0.6
Period 2	1.62	1.85	1.90	0.12	0.15	0.83
Period 3	1.65	1.66	1.72	0.12	0.76	0.72
	DMI, kg/d
Total	4.15	4.20	4.22	0.12	0.68	0.91
Period 1	3.83	3.85	3.96	0.11	0.59	0.53
Period 2	4.28	4.35	4.38	0.15	0.65	0.86
Period 3	4.34	4.41	4.32	0.12	0.87	0.61
	G:F, kg/kg
Total	0.36	0.38	0.40	0.01	0.07	0.53
Period 1	0.34	0.37	0.38	0.02	0.38	0.77
Period 2	0.38 a	0.43 b	0.43 b	0.02	0.05	0.83
Period 3	0.36	0.36	0.37	0.02	0.63	0.41

^a,b Within a row means without common superscripts differ, p < 0.05 by MIXED procedure of SAS 9.4. The experimental design and statistical analysis of this feedlot study was performed as described in Materials and Methods. ¹ ADG, average daily gain; DMI, dry matter intake; G:F, gain to feed.

. There were no dietary treatment effects for overall ADG or DMI at any period. However, the G:F tended to be greater (p < 0.07) for the average of the MON and THYM groups, compared to the CON group (0.38 and 0.40 vs. 0.36 kg/kg, respectively), over the 63-day trial and was significantly higher (p < 0.05) than the CON group during period two (0.43 and 0.43 vs. 0.38 kg/kg, respectively). The inclusion of THYM or monensin in the diet did not reduce feed intake.

3.4. Amount of XN and Related Phytoestrogens in Serum of THYM-Fed Animals

Given the potential for XN, a prominent hop metabolite, to be converted to the phytoestrogen 8-PN [46], we used a highly sensitive and specific method of LC-MS-MS [45] to measure XN metabolites in serum from animals fed THYM. The results of this analysis (

Table 6. Concentration of XN and metabolites in the serum from steers in the 1% THYM treatment group ^1,2.

Metabolite 3	THYM	SD
	Serum Concentration (nM)
XN 2	7.3	1.7
DXN	5.2	1.4
IXN	0.5	0.3
6PN	6.9	1.4
8-PN	4.0	0.8
DDXN	1.0	0.3

¹ Serum samples were collected at day 63 from steers (N = 8) on TMR with 1% THYM (THYM group) in Table 5. ² Metabolites were undetectable in serum samples of steers (N = 4) on TMR without THYM (CON group) in Table 5. ³ XN, xanthohumol; DXN, α,β-dihydroxanthohumol: IXN, isoxanthohumol; 6PN, 6-prenylnaringenin; 8-PN 8-prenylnaringenin; DDXN, O-desmethyl-α,β-dihydroxanthohumol.

) show nanomolar levels of XN and its metabolites in the serum from blood taken from animals supplemented with 1% THYM. These values were similar between the THYM animals, as shown by the low standard deviations. No XN metabolites were found in control animals, nor was 8-PN seen in THYM.

4. Discussion

4.1. Origin and Analysis of THYM

The cattle industry already utilizes spent brewer’s grain as a feedstuff, but remaining byproducts of craft brewing (trub, hop acids, and spent yeast) are typically discarded as waste. THYM, with α and β hop acid content in the range of 22–30 mg/g and 14–17 mg/g respectively, as shown in Table 3, contains hop acid concentrations at a magnitude greater than those reported for spent craft yeast alone [13]. This likely reflects the hop acid contributions of trub and spent hops as shown in Figure 1A. As noted in Figure 1B, hop acids are hydrophobic, indicated by their high log p values, and are poorly soluble in water. Hence, most of these hop metabolites end up in the solids associated with THYM. The high concentration of hop metabolites, including in THYM, particularly the presence of 1–2 mg/g of XN, coupled with its well-explored biological and medicinal properties [3], highlights the potential value of THYM as a feed supplement. Our work further found that drum drying, an energy efficient process [47], could quickly reduce the water content of THYM from >90% to <10%, with >75% retention of hop acids.

This drying process greatly enhances the attractiveness of THYM, and the drum drying process itself has attracted attention for further efficiency advances [48]. Other THYM dewatering approaches, such as drying filter cakes or spray drying, were not as effective as drum drying [39].

4.2. Antimicrobial Activity of Hop Acids and THYM

THYM’s antimicrobial properties are substantial, shown through clear zones of growth inhibition on agar diffusion (Figure 3) and by microtiter dilution (Table 1) where a MIC of 137 ± 39 3 μg/mL was observed. The growth inhibition of α-acids, β-acids, and monensin of 8, 2, and 3 μg/mL are consistent with the literature for B. subtillis [49,50,51,52]. Βeta-acids also inhibited Listeria monocytogenes at 3 μg/mL [53] and the equine fecal bacteria Streptococcus bovis, with a MIC of 18–45 μg/mL [54]. Furthermore, hop extracts have shown antimicrobial activity on medically relevant organisms such as Clostridioides difficile (C. dif.) [55] and biofilms [56]. The prenylated polyphenol, XN, likely also contributes to the antimicrobial activity of hop extracts [16]. Though limited to a few strains, these in vitro values demonstrate the general susceptibility of Gram-positive bacteria to β-acids. This assertion is further supported by a report that the addition of a hop extract to in vitro batch fermentations with mixed human fecal inoculum altered the microbiota profile at a β-acid concentration of 29 μg/mL [57]. Furthermore, hop β-acids and extracts have shown promise in clinical use against several Gram-positive pathogens [16,55,58], further attesting to their significant potency. While monensin is generally considered to be a more potent antimicrobial than hop β-acids [20], there are some findings highlighting a somewhat comparable efficacy. A comparison of a hop extract with β-acids at 23 μg/mL and monensin at 2.5 μg/mL resulted in a comparable reduction in methane and ammonia N, and increased propionate in rumen fermentation studies [59].

4.2.1. Antioxidant Activity of Hop Acids and THYM

Our data showing greater antioxidant capacity for α vs. β hop acids agree with earlier studies on the antioxidant activity of hop components [18]. The TEAC antioxidant activity, 73 μmol/g of the 70% ethanol:30% water THYM extract, compared favorably with vitamin C at 494 μmol/g (Table 1). However, unlike highly water-soluble vitamin C, only 7% of antioxidant in THYM was extractable with water. Further analysis of THYM’s antioxidant activity would benefit by measuring biomarkers of oxidative damage in vivo. In particular, the lipid oxidation biomarkers, malondialdehyde (MDA) and isoprostanes [60] would be particularly useful given the hydrophobic nature of THYM’s antioxidant activity. Notably, hydrophobic hop acids are likely to exert antioxidant activity in lipid-rich membranes [61].

4.2.2. Rumen Fluid Fermentations with THYM

The inhibition of rumen fluid methane production by 1.3–3.3 mg/mL THYM, in Table 2, replicate earlier reports οf rumen bovine fluid fermentations where addition of 0.8 mg/mL of powdered hop pellets reduced methane and the A/P ratio [26]. For both experiments, the calculated β-acid concentration was approximately 60 μg/mL. The results shown in Table 2 were replicated in similar batch fermentations using bovine rumen fluid obtained after slaughter where THYM at 4 mg/mL and monensin at 10 μM inhibited methane formation and reduced the A:P ratio to a similar extent (R. Bryant and L. Martin, unpublished observations).

While the possible mechanism of hop acid inhibition of methane is unclear, β-acids were found to inhibit methane production by Methanobrevibacter ruminantium [62]. While hop acids in THYM are likely the major factor causing the observed changes in SCFA and methane metabolism, the significant amounts of protein found in THYM could also influence the incubations. However, added protein played little role in methane inhibition in other experiments where hop acid-rich spent craft yeast was compared with protein-rich baker’s yeast [23]. That said, in some incubations, THYM represented 10% of the added substrate and non-hop acid components, e.g., crude protein, could have influenced the in vitro metabolism, as could have the pH drop from 6.8 to 5.4 over the 24 h fermentation period.

It should be noted that our fermentation metabolism data were obtained via an in vitro batch mode, only partially mimicking in vivo conditions [63]. This method coupled with a one 24 h time point does not capture temporal dynamics, removal of substrate, buildup of gases, nor the rhythmic contractions of the rumen. Demonstrating the usefulness of THYM as a suppressor of rumen methane formations will require studies with live animals [44,64].

4.3. In Vivo Feedlot Studies

4.3.1. Selection of the THYM Inclusion Rate

The selection of 1% THYM as the feed inclusion rate was influenced by the presence of bitter iso- α-acids [65] and other possibly bitter phytochemicals [66]; these raised the possibility that high levels of THYM, e.g., 10%, could reduce intake. Another consideration was the comparable potency, within an order of magnitude, between hop β-acids and monensin. At 1% THYM inclusion, rations would have 150 mg/kg β-acids for the THYM group, an amount 5-fold higher than the inclusion rate of monensin at 30 mg/kg in the MON group. Furthermore, 1% THYM would contribute 250 mg/kg of α-acids, which also showed strong antimicrobial activity (Table 1).

4.3.2. Animal Performance with THYM and Monensin Diet Inclusion

The feed efficiency results over the 9 weeks for the THYM and monensin groups tended to be higher (p < 0.07) when contrasted to the control group. THYM and monensin showed improved feed efficiency in period 2 (Table 5). However, the performance for both monensin and THYM was lower than expected, perhaps because of a high-feed forage content [21,26,27].

Overall, this 63-day feeding trial suggested productive differences in feed efficiency, with the THYM and monensin groups tending to gain more per kg of feed throughout the trial. The gains per kg of feed were significant between day 21 and 42. It should be noted though that these in vivo results should be seen as an early, short-term study in one strain and sex and need to be confirmed with further research under more conditions.

THYM had no apparent negative effects on dry matter intake when used as a supplement. The presumed bitterness of these brewing byproducts did not lead to feed rejection. Its strong response in the ABTS antioxidant assay suggested that the incorporation of 1% THYM could provide an antioxidant potential to feed. Furthermore, its strong in vitro Gram-positive antimicrobial activity and in vitro activity on mixed rumen microbes is supportive of exhibiting in vivo activity. The 37% crude protein content of THYM, mostly of yeast and grain origin, is also of nutritional value and could be a source of immune-beneficial β-glucans and mannans [67]. It should be noted that the final calculated crude protein (CP) value for the THYM diet was lower than the CON and MON diets by a factor of 4%.

Inclusion of hops or purified hop acids in cattle feeding trials has suggested performance benefit in some instances. Addition of ground hop pellets to feed (0.05% and 0.1%) in a growth and finishing feedlot 22-week study suggested a trend of a 6% increase in ADG versus the control diet [28]. The contents in rations of the hop acids in this study [28], 4 and 7 mg/kg for α-acids and 42 and 84 mg/kg for β-acids, were less than in our study where the content of α-acids was 254 mg/kg and average β-acids content was 154 mg/kg. On the other hand, a 9-week cattle growth study utilizing either 0.5% or 1% dry hop cones in rations did not alter weight gain or feed efficiency in Cika bulls [29]. The diet contents of hop acids in this study, 633 and 1265 mg/kg α-acids and 219 and 539 mg/kg β-acids, were higher than in our study.

As noted above, hops [28,29] and hop extracts have been studied as feed supplements [68], but their cost could be an impediment to widespread use [21]. Given that the mixture THYM is a recovered waste product, it may be the more economical way of adding valuable hop-derived phytochemicals to livestock feed. THYM represents a way of upcycling a portion of the annual hop harvest. This is particularly true for the antimicrobial and antioxidant hydrophobic hop metabolites that would otherwise be lost.

4.3.3. Estrogenic Evaluation of THYM

In addition to antimicrobial and antioxidant activity, hops are noted for their estrogenic potential [69]. Since concerns have been raised about phytoestrogenic effects in cattle [70,71], we chose to measure the amounts of 8-phenynaringenin (8-PN) in the serum of animals supplemented with THYM. XN metabolites are potent estrogens and, while our feedlot study was performed in steers, we thought measuring blood levels of these bioactive molecules would be useful. It should be noted that XN metabolites largely result from gut metabolism of XN [72]. Very little 8-PN was found in THYM samples (W. Wu, unpublished observations). The blood level of 8-PN was 4 ± 1 nM in the THYM group, with an estimated XN intake of 0.2 mg/kg BW/day.

The blood level of 4 nM 8-PN in THYM-supplemented steers, were it in female cattle, would seem unlikely to trigger major alterations to reproductive functions. For perspective, a study was done with different cattle herds that compared blood levels of phytoestrogenic isoflavones to incidence of silent heat (hidden estrus), an example of reduced reproductive performance. The herd with the highest level, 4.9 ± 0.5 μM of phytoestrogens, >1000 times concentration of 8-PN observed in our study, had a 41% incidence of silent heat while the herd with lowest level, 1.0 ± 0.03 μM, had a 7% incidence [70]. In a study of estrogenic-driven uterine weight gain, rats given amounts of 1260 mg/kg of 8-PN in feed had an 8-PN serum level of 109.8 μM and exhibited uterine weight gain. Nonetheless, rats on a 10-fold lower supplementation of 126 mg/kg, with 8-PN blood levels of 14.5 uM, >3000 times 4 nM, experienced no effect on uterine weight [73]. Even though the estrogenic potential of THYM seems low, reproductive physiology studies would be helpful in ruling out risks. All the same, it should be note that while high levels of phytoestrogens can trigger changes in reproductive function, low levels can enhance growth [71]. Recent research has also demonstrated that XN and its hydrogenated α,β-dihydro-XN (DXN) and tetrahydro-XN metabolites have antiproliferative effects on colon cancer cells [74] and improve gut function [75].

More approaches are needed for sustainable food production, management of waste streams, and reduction in greenhouse gas emissions [76]. Our research on brewing byproducts, emphasizing the presence and value of hop-derived natural antimicrobials and antioxidants [77,78,79], suggests that directly upcycling the underutilized brewing byproducts, namely spent yeast, hops, and trub, follows this path. They are rich in antimicrobial and antioxidant phytochemicals [80], which have demonstrated positive metabolic and nutritional properties [18,81]. Of particular interest is a recent study with nursery pigs fed THYM at several inclusion rates, where there was an increase in feed efficiency, an apparent increase in rate of intestinal tissue repair, and an increase in the expression of genes related to innate immune system recognition and signaling without causing the elevation of inflammatory responses [82]. Capturing and upcycling the waste streams of trub, hops, and yeast will advance the above sustainability objectives and could provide needed livestock growth promotion alternatives.

In addition, the craft brewing industry produces millions of tons of biological waste annually. Disposing of trub, hops, and yeast is conventionally expensive for breweries and environmentally damaging due to its high biological oxygen demand (BOD) in water systems. The upcycling of trub, spent hops, and yeast into a valuable product prevents environmental degradation by diverting high-load organic waste from landfills, lagoons and municipal water treatment facilities. For instance, the Asheville Municipal Sewer District in Asheville, NC, USA, receives wastewater from approximately 140 craft breweries, cideries and distilleries, and is interested in finding ways to reduce loading from the brewing industry.

5. Conclusions

Beer brewing is a complex process accompanied by the large generation of several byproduct streams that must be managed and disposed of sustainably to avoid environmental impacts. A process was found to collect and dry a feed supplement from the underutilized brewing byproducts of trub, spent hops, and yeast. An important aspect of this collection process is that significant amounts of the antimicrobial and antioxidant α and β hop acids and XN, initially present in the added hops and a costly ingredient for brewers, are recovered in the mixture of byproducts, giving it added value. Furthermore, removal of these nutrient-rich materials from brewing waste streams will lessen the burden on municipal water treatment facilities and enhance brewery sustainability.