The essential oil blend Agolin Ruminant L reduces methane production in vitro and in vivo when included in the drinking water of cattle

Abstract

Two experiments were conducted to determine the potential for the essential oil blend Agolin Ruminant L (Agolin) to reduce enteric methane (CH₄) emissions from beef cattle when delivered via drinking water. Experiment 1 evaluated aqueous solutions of Agolin (50 mg/L) and a nonprotein nitrogen and mineral solution (uPRO ORANGE [uPRO]; 1.7 mL/L) individually and in combination, where Agolin was added to concentrated uPRO at 3%, 4.5%, and 6% Agolin (w/w) prior to dilution with water at 1.7 mL/L, for a total of 5 treatments. These were incubated for 48 h with a medium-quality Rhodes grass (Chloris gayana) hay substrate, with gas production, CH₄ concentration in gas, and digestibility measured in vitro. In experiment 2, Droughtmaster steers (n = 24) were fed a basal diet of Rhodes grass hay and were allocated to 1 of 3 water treatments (n = 8 per treatment) supplemented with either uPRO (2.27 mL uPRO/L water), or 1 of 2 inclusion rates of Agolin in combination with uPRO (2.27 mL uPRO and 6 µL Agolin/L water or 2.27 mL uPRO and 24 µL Agolin/L water) with enteric CH₄ emissions, feed and water intake, and live-weight gain (LWG) measured over 56 d. In experiment 1, the inclusion of Agolin in uPRO at 6% (w/w) resulted in a reduction in CH₄ production (15.8%; P = 0.003) and the proportion of CH₄ in the gas produced (24.5%; P < 0.001). In experiment 2, steers consuming the lower quantity of Agolin via drinking water had a 16.4% (P = 0.0027) reduction in CH₄ production over the experiment, declining from 140 g/d during week 1 to 117 g/d in week 8. This inclusion rate of Agolin in the drinking water also resulted in a 25 g (17.6%) CH₄/d decrease in emissions by steers compared to control steers (P = 0.0205). However, no significant differences in CH₄ yield (g CH₄/kg dry matter intake), or CH₄ intensity (g CH₄/kg LWG) by steers were observed between treatments. These results demonstrated that Agolin reduces CH₄ emissions when mixed in an aqueous solution under in vitro and in vivo conditions, providing a potential method to reduce enteric CH₄ emissions from cattle in extensive production systems.

Delivery of methane-reducing compounds to beef cattle in extensive production systems presents unique challenges. The potential to use drinking water to facilitate the delivery of such compounds to cattle was investigated.

Introduction

Global beef production faces a challenge to increase production to meet future demand, whilst also addressing climate change concerns linked to the livestock sector (Beauchemin et al., 2020). Of all greenhouse gases generated by the livestock sector, approximately 39% are associated with methane (CH₄) released during enteric fermentation of feedstuffs by ruminants (Gerber et al., 2013).

The locations of Australia’s beef production systems are diverse, ranging from the extensive northern tropical regions to the more intensive southern temperate regions (Greenwood, 2021). The predominantly pasture-based systems that characterize much of the country’s red meat industry also experience seasonal variations in forage production and nutritional quality, which differ between the temperate and tropical regions (Costa et al., 2012). Supplementation strategies ensure the provision of limiting nutrients to animals, which may be aimed at minimizing livestock losses in extensive situations, but in other production systems are aimed at increasing live-weight (LW) gain (LWG; de Oliveira et al., 2022). The inclusion of CH₄-reducing compounds in nutritional supplements presents an opportunity to deliver CH₄-suppressing compounds to animals in extensive production systems.

Several CH₄-reducing additives have been developed for the livestock industries; however, they have primarily been targeted to, and evaluated in dairy or feedlot applications (Beauchemin et al., 2022) where feed delivery and intake are tightly controlled. These existing delivery methods are not generally suited to most of Australia’s extensive pasture-based production systems. In these systems, consistent delivery of additives is a challenge due to remoteness, environment, infrastructure, and animal behavior. Common to both intensive and extensive production systems, however, is the requirement of animals for consistent access to drinking water, often via a reticulated system. Delivery of CH₄-reducing compounds and other supplements through drinking water troughs may provide an innovative mechanism to reduce CH₄ emissions on a scale not possible with existing techniques.

In the context of exploring water-delivered antimethanogenic supplements, the essential oil blend Agolin Ruminant L (Agolin; Alltech Technology, Nicholasville, KY) stands out as a promising candidate given its demonstrated results in dairy cattle (Belanche et al., 2020), existing approval for use in ruminants, and water-soluble liquid form. The objective of the current experiments was to determine the efficacy of Agolin for reducing enteric CH₄ emissions in beef cattle when delivered in conjunction with the nonprotein nitrogen (NPN) and mineral supplement uPRO ORANGE (uPRO; DIT Ag-Tech, QLD, Australia) via the drinking water. The incorporation of a nutrient supplement in the study was based on the rationale that producers are likely to utilize such a supplement to target common nutrient deficiencies, and Agolin could be included as a secondary additive. The study also sought to quantify the effect of consumption of Agolin via drinking water on productivity of beef cattle. It was hypothesized that the delivery of Agolin to cattle via drinking water would decrease enteric CH₄ emissions without negatively impacting rumen fermentation, productivity, or blood parameters.

Materials and Methods

The use of animals in this study was conducted in accordance with the guidelines of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes and was approved by the CQUniversity Animal Ethics Committee (24410).

Experiment 1

A single in vitro fermentation experiment was conducted following a modification to methods described by Kinley et al. (2016) and Schlau et al. (2021). Briefly, approximately 0.5 g of Rhodes (Chloris gayana) grass hay (931 g organic matter [OM], 37 g crude protein [CP], and 699 g neutral detergent fiber [NDF]/kg dry matter [DM], 2 mm particle size) was added to an F-57 filter bag (Ankom Technology, Macedon, NY) previously treated with acetone, dried, and weighed as per the manufacturer’s instructions. After sealing, the top portion of the bag was sliced and a 1-cm long, 6-mm stainless steel bolt was inserted and secured with the accompanying nut to ensure the bag remained suspended in the inoculum. Each filter bag was placed inside a 250-mL septa port vessel (Simax, Křížová, Prague 5, Czech Republic) with 100 mL of prewarmed buffer (Goering and Van Soest, 1970) and 20 mL of prewarmed water (control, blanks), Agolin or uPRO, or 3 increasing amounts of Agolin combined with uPRO added to vessels (Table 1) in triplicate. The Agolin concentration in the treatment solutions was designed to reflect 622, 933, and 1,244 mg Agolin delivered to a 300-kg LW steer consuming 14 L of water per day. Concentrations were chosen based on previous in vitro work in our laboratory (Batley et al., 2024a, 2024b), and existing literature (Belanche et al., 2020), which indicated the minimal impact of Agolin on CH₄ production at lower concentrations in vitro.

Table 1.

Concentrations and simulated delivery rates of treatments evaluated in vitro (experiment 1)

Treatment	Abbreviation	Simulated Agolin delivery (mg/animal/d)
Rhodes grass control	RGC	0
Agolin Ruminant L1, 50 mg/L	AR	700
uPRO ORANGE2 w/w, 1.7 mL/L	UO	0
3% Agolin Ruminant L/97% uPRO ORANGE w/w, 1.7 mL/L	3 %	622
4.5% Agolin Ruminant L/95.5% uPRO ORANGE w/w, 1.7 mL/L	4.5%	933
6% Agolin Ruminant L/94% uPRO ORANGE w/w, 1.7 mL/L	6 %	1,244

¹Agolin Ruminant L (Alltech Technology) from plant extracts, including coriander (Coriandrum sativum) seed oil (up to 10%), eugenol (up to 7%), geranyl acetate (up to 7%), and geraniol (up to 6%) with preservatives such as fumaric acid (Belanche et al., 2020).

²uPRO ORANGE (DIT Ag-Tech, QLD, Australia) is composed of 19.79% nitrogen, 4.28% phosphorus, and 1.20% sulfur.

The vessels were then placed in 2 shaking water baths maintained at 39 °C for 30 min before the addition of 2 mL of a prewarmed reducing solution and 25 mL of well-mixed rumen fluid collected (Batley et al., 2022) from 7 grass-fed Brahman and Charolais cross steers (465.3 ± 22.0 kg LW mean and SD) slaughtered at an accredited abattoir (JBS Australia, Rockhampton, QLD, Australia). Briefly, samples of fluid were collected from various rumen locations and filtered through 4 layers of cheesecloth into a preheated thermos-style flask. Samples of rumen mat were also taken and squeezed through the cheesecloth. The headspace in the flask was flushed with CO₂, and fluid was transported to the laboratory within 2 h of collection. Upon laboratory arrival, the rumen fluid was transferred to a Schott bottle, the headspace flushed with CO₂, then topped with a 50-mL bottle top dispenser (Fortuna Optifix, Poulten & Graf, Wertheim, Germany) calibrated to dispense 25 mL. Fluid was maintained at 39 °C in the water bath and mixed before inoculation of each fermentation vessel. The vessels were sealed with Ankom RF gas production modules, and CO₂ flushing was conducted as per the manufacturer’s instructions. The water baths were set to oscillate at 85 rpm with fermentation proceeding for 48 h. The parameters for the Ankom RF gas production unit were configured as follows: maximum pressure = 3 psi, live interval = 60 s, recording interval = 20 min, and valve open time = 250 ms. At the end of the fermentation, vessels were placed on ice to minimize microbial activity.

Sample collection and CH₄ measurement

Upon completion of the fermentation, 20 mL of gas was sampled from the headspace with a gas-tight syringe from the side port on the reaction vessel. This sample was transferred to an evacuated 12 mL Exetainer vial (Labco, Lampeter, Ceredigion, Wales) for CH₄ analysis using gas chromatography, completed on an Agilent 6890N gas chromatograph (Santa Clara, CA) fitted with a Supelco (Bellefonte, PA) 80/100 HayeSep Q 3FT × 1/8 IN × 2.1 mm stainless steel column and a flame ionization detector. Methane concentration was combined with gas pressure data to calculate the total gas and CH₄ produced during fermentation, reported as milliliter per gram of the substrate. A 4-mL subsample of the fermentation fluid was added to 1 mL of 20% metaphosphoric acid with internal standard (11 mM 4-methylvaleric acid) and stored at −20 °C prior to volatile fatty acid (VFA) analysis. The Ankom bags were removed and retained to determine in vitro DM digestibility (IVDMD) and in vitro OM digestibility (IVOMD).

Experiment 2

Experimental design

The 56-d long experiment was conducted at CQUniversity (Rockhampton, QLD, Australia). Droughtmaster steers (n = 24; 263.9 ± 22.9 kg LW mean and SD) with radio-frequency identification (RFID) tags (NLIS cattle tags; Allflex, Murarrie, QLD, Australia) in the left ear were ranked and blocked on LW and allocated to 1 of 3 treatments (n = 8 steers per treatment) within each LW block. The treatments were the addition of uPRO only (T1) or uPRO with low (T2) and high (T3) doses of Agolin added to the drinking water available for consumption by the steers (Table 2).

Table 2.

Description of experimental treatments, target and average actual Agolin delivery rate, and pH of drinking water available to steers in experiment 2

Treatment	Treatment description	Agolin delivery (mg/animal/d)		Drinking water pH3
		Target	Actual
T1	uPRO ORANGE1, 2.27 mL/L	0	0	3.04 ± 0.12
T2	uPRO ORANGE, 2.27 mL/L; Agolin Ruminant L2, 6 µL/L	100	107.4	3.17 ± 0.21
T3	uPRO ORANGE, 2.27 mL/L; Agolin Ruminant L, 24 µL/L	400	434.5	3.33 ± 0.50

¹uPRO ORANGE (DIT Ag-Tech, QLD, Australia) is composed of 19.79% nitrogen, 4.28% phosphorus, and 1.20% sulfur.

²Agolin Ruminant L (Alltech Technology) from plant extracts, including coriander (Coriandrum sativum) seed oil (up to 10%), eugenol (up to 7%), geranyl acetate (up to 7%), and geraniol (up to 6%) with preservatives such as fumaric acid (Belanche et al., 2020).

³Mean and SD of 2 pH measurements of each water treatment each day over the 56 d experimental period.

Steers had ad libitum access to Rhodes grass hay (910 g OM, 51 g CP, and 710 g NDF/kg DM), chaffed to approximately 50 mm in length, via 7 smart feeders (SmartFeed Pro, C-Lock Inc., Rapid City, SD). Fresh hay was provided twice daily at approximately 0800 and 1500 hours. Water treatments were also available ad libitum via 3 smart feeders of the same make and model used to provide the hay. The only modification to these feeders was the use of an additional plastic tub of 100 L capacity placed inside the feeder to hold the water. Access of steers to their allocated water treatment was controlled via the unique RFID ear tag, but steers could access the Rhodes grass hay from any feeder. Steers were maintained as a single group throughout the experiment in 2 under-cover, concrete-floored (180 m² each) pens with rubber mats connected by an outdoor, dirt-floored, open-air loafing area (~99 m²). There was approximately 19 m² of floor space available per steer (Figure 1).

Figure 1.

Schematic of in vivo trial area. A secured central feed area contained the SmartFeed Pro feeders, which allowed access for cleaning, filling, calibrations, and maintenance. This central area was straddled by 2 group pens, which, along with the feed area, were covered. Feeders not marked as water treatments denote those dedicated to hay access. The location of the GreenFeed unit is shown in dark green. Except for the secured central feed area, animals had access to all areas at all times.

All feeders were monitored daily using the C-Lock mobile application and were also verified weekly and recalibrated as required following the manufacturer’s guidelines. Prior to the commencement of the experiment, a 2-wk training period was implemented to ensure steers could access the feeders successfully. During this time, the feeders were set to training mode to allow steers to access any feeder, and the feeders used to supply water were filled with water supplemented with uPRO only. After this training period, water in the feeders was supplemented with Agolin in the concentrations noted earlier. Due to an RFID tag issue, one steer from the T1 treatment was removed from the experiment.

Water was supplied via an automatic uDOSE system (DIT Ag-Tech, Wilsonton, Queensland) and sent through pipelines to the feeders. Each supplement had a dedicated uDOSE module, which automatically added 10 mL of a preprepared concentrated supplement base/L of water supplied to achieve target delivery rates. The DIT Ag-Tech dosing system reported that the target delivery rate of supplements was achieved at a rate of 90.5%. The water delivered in the feeders was provided at fixed times (0800, 1300, and 1600 hours) to ensure adequate supply to the steers. A pH measurement was taken each time water was added to the feeder as a means of further confirming the delivery of the uPRO and Agolin into the drinking water.

Measurements

Daily feed and water intake of individual animals were measured using the auto-feeders with data provided daily as a report emailed from C-Lock. Steers were weighed prior to the morning feeding every 7 d during the experimental period.

Measurement data for enteric CH₄, carbon dioxide (CO₂), and hydrogen (H₂) emissions were collected throughout the experiment using the GreenFeed emissions monitoring (GEM) system (C-Lock Inc.) located in the outdoor connecting loafing area described above and as per the manufacturer instructions (Figure 1). Access to the GEM was verified (3.55 ± 1.27 min visit duration mean and SD) through an on-animal RFID ear tag, and data were accessed weekly via the C-lock web portal. Data from visits under 2 min duration were not included.

Steers were attracted to the GEM unit with pellets (Barastoc Calm Performer, Ridley Corporation Ltd, Melbourne, VIC, Australia; 875 g DM, 170 g CP, 140 g crude fiber, 30 g ash, 30 g ether extract, 12 g Ca, and 6 g P/kg DM). The GEM unit was programmed to allow up to 5 feeding events per animal per day, with a minimum interval of 4 h between each visit. During the training phase, a maximum of 8 pellet drops (35 g/drop) were permitted for each feeding visit, and after adaptation, the drops were reduced to 4 per visit with an interval of 35 s between each drop. The GEM was calibrated at the beginning and end of the trial, with further CO₂ recoveries performed every 30 d to ensure the accuracy of the gas sensor readings. Methane production was reported as g CH₄/d, CH₄ yield was reported g CH₄/kg DM intake (DMI), CH₄ intensity was reported as g CH₄/kg LWG, and the ratio of CH₄ to water intake was also reported as g CH₄/L of water intake.

Sample collection

Subsamples of feed offered were collected during the preparation of daily feed allowances and bulked over 7 consecutive days. Feed refusals were collected from each feeder prior to feeding each morning and were also subsampled and bulked over 7 consecutive days. Bulked weekly subsamples of feed offered and refused were mixed thoroughly with duplicate subsamples dried to a constant mass at 60 °C prior to analysis for DM and neutral detergent fiber digestibility (NDFD).

Fecal samples were collected from the rectum of each steer at 0800 hours on days 50 to 54 of the experiment and dried to a constant mass at 60 °C. A sample of the feed offered during that period was also collected and dried in the same manner. Dried feces and feed were homogenized and ground to a 2-mm particle size for NDFD analysis.

On day 56 of the experiment, blood and rumen fluid samples were collected from the steers. Blood samples were collected from the jugular vein of the steers into lithium heparin-coated vacutainers (Becton Dickinson, Franklin Lakes, NJ). Vacutainers were placed on ice for approximately 1 h before centrifugation at 2,000 × g at 4 ° for 15 min with plasma removed and stored at −20 °C prior to analysis. Rumen fluid was collected using a stomach tube attached to a 2-way hand pump and strained through nylon fabric as described by Parra et al. (2021). A 4-mL subsample of rumen fluid was added to 1 mL of the same acid and spiking solution noted in experiment 1, before being stored at −20 °C prior to VFA analysis.

Analytical

Crude protein

Nitrogen content of the feed substrate was determined using a LECO TruMac Series Carbon and Nitrogen Analyzer (Sydney, Australia) and converted to protein content using a conversion factor of 6.25 (Sweeney, 1989).

Neutral detergent fiber

NDF was measured using an Ankom (Macedon, NY) model 200 fiber analyzer as per the manufacturer’s instructions, adapted from Van Soest et al. (1991).

Digestibility

Apparent IVDMD and IVOMD of the Rhodes grass hay substrate in experiment 1 were determined using the Ankom bags at the end of the 48 h fermentation period. Bags were rinsed with cold water until the water became clear, and then dried to constant mass at 105 °C to determine IVDMD. Following this, the bags were incinerated at 550 °C for 8 h with the ash weighed to determine IVOMD. Drying and furnacing protocols were based on Kinley et al. (2016).

In experiment 2, NDFD was estimated using indigestible NDF (iNDF) as an internal marker, following a slightly modified method to that described by Parra et al. (2021). Briefly, homogenized feed and fecal samples were weighed (0.5 g) into F-57 filter bags in triplicate and digested for 240 h in a Daisy Incubator (D200, Ankom Technology). The rumen fluid used was collected from the same abattoir using the same procedure described in experiment 1 but from different steers, and was replaced with fresh rumen fluid after 120 h. The first rumen fluid sample came from 7 Brahman cross steers (604.4 ± 19.9 kg LW mean and SD), and the second from 3 Santa Gertrudis cross steers (465.2 ± 4.2 kg LW mean and SD). After the fermentation, bags were rinsed and analyzed for NDF content with NDFD determined using the following equation:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{N}\mathrm{D}\mathrm{F}\mathrm{D}=100-[100\times \left(\frac{\mathrm{D}\mathrm{i}\mathrm{e}\mathrm{t}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\left(\mathrm{\%}\right)}{\mathrm{F}\mathrm{a}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{N}\mathrm{D}\mathrm{F}\left(\mathrm{\%}\right)}\right)\times \left(\frac{\mathrm{F}\mathrm{a}\mathrm{e}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{N}\mathrm{D}\mathrm{F}\left(\mathrm{\%}\right)}{\mathrm{D}\mathrm{i}\mathrm{e}\mathrm{t}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{N}\mathrm{D}\mathrm{F}\left(\mathrm{\%}\right)}\right)].}\end{array}}$$

Biochemical analytes in plasma

The concentration of blood parameters was assessed using an Olympus AU400 autoanalyzer (Beckman Coulter Inc., Melville, NY), following the procedure outlined by Costa et al. (2016), with a Beckman Coulter Diagnostic Systems kit and reagents. The analytes measured included sodium, potassium, chloride, bicarbonate, glucose, urea, creatinine, calcium, phosphate, magnesium, total protein, albumin, creatine kinase, aspartate aminotransferase, glutamate dehydrogenase, gamma-glutamyl transferase, bilirubin, bile acids, triglycerides, beta-hydroxybutyrate, and nonesterified fatty acids. Only parameters that differed significantly were reported in the results, except for plasma urea nitrogen, phosphate, and glucose, which were reported regardless of significance.

Volatile fatty acids in rumen fluid

Thawed rumen fluid samples were centrifuged at 2,000 × g at 4 °C for 20 min, and approximately 1.5 mL of the supernatant was passed through a 0.22-µm syringe filter into a sample vial. Samples were analyzed for VFA content using gas chromatography–mass spectrometry (Shimadzu QP2010 Plus; Nakagyoku, Kyoto, Japan) and an Agilent (Santa Clara, CA) HP-INNOWAX column (30 m × 0.25 mm × 0.25 µm; Batley et al., 2024a). A multiacid standard was used in combination with the internal standard for calibration curves and concentration calculations.

Statistical analysis

For experiment 1, individual fermentation vessels were allocated as the experimental unit. Where analysis of variance results are shown, and analysis was completed using IBM SPSS (v26). A threshold for significance of P-value < 0.05 was used. Where applicable, results are presented as the mean ± standard error of the mean.

In experiment 2, individual animals were allocated as the experimental unit. The repeated measures method was applied to verify the effect across all experimental weeks for the following reported variables: DMI, LWG, water intake, GEM measurements, CH₄ production per yields, H₂, and CO₂. For these variables, the model applied was:

$${\displaystyle \begin{array}{l}{\displaystyle Y_{ij}=\mathrm{\mu }+{\mathrm{\tau }}_i+{\mathrm{\beta }}_j+{(\tau \mathrm{\beta })}_{ij}+{ϵ}_{ij}}\end{array}}$$

where Y_ij is the observation for the ith treatment at the jth time point. μ is the overall mean, τ_i is the effect of the ith treatment, β_j is the effect of the jth time point; (τβ)_ij is the interaction between the ith treatment and the jth time point; and ϵ_ij is the random error associated with the observation for the ith individual at the jth time point.

Rumen (VFA, pH) and plasma parameters, CH₄ intensity, and NDFD were analyzed as a completely randomized design. The model for these analyses was:

$Y_{ij}=\mathrm{\mu }+{\mathrm{\tau }}_i+{\in }_{ij},$

where Y_ij is the observation from the ith treatment (factor level) on the jth experimental unit. μ is the overall mean, τ_i is the effect of the ith treatment, and ϵ_ij is the random error, assumed to be independent and identically distributed normal with mean zero and constant variance.

The Kolmogorov–Smirnov test for normality was utilized for all variables, being considered normal when P-value > 0.05. When non-normal, raw data were transformed to fit a normal distribution prior to analysis. Variables for CH₄ yield as a quotient of DMI, ratio of CH₄ to water intake, DMI, pellets consumed, hydrogen production, and total feed intake were treated in this manner. Pellets consumed was the only variable unable to be transformed to fit a normal distribution (P = 0.03).

Raw data from the GEM underwent quality control prior to statistical analysis. The mean and SD were calculated for each variable (CH₄, H₂, and CO₂) from each animal during each week of the experiment, with outliers beyond ±2SD being discarded from the data set. Following this, data were analyzed following a completely randomized design with repeated measures, and the interaction was considered in the model for all these variables. The residuals were modeled considering the covariance between repeated measures of the same animal at the time intervals by employing different residual (co)variance structures (corAR1, corARMA, corCAR1, corCompSymm, corExp, corGaus, corLin, corRatio, corSpher, and corSymm), and the best structures were chosen based on Akaikes information criterion. For further comparison between treatments, paired differences were used. All statistical analyses for experiment 2 were performed using R (version 4.3.2, RStudio, Boston, MA, 2023) with a threshold for significance of P-value < 0.05.

Results

Experiment 1

The combination of Agolin included at 4.5% with uPRO increased (P = 0.006) total gas production in vitro above that when Agolin and uPRO were used alone or were excluded from the fermentation completely (Table 3). The concentration of CH₄ in total gas produced was lower (P = 0.003) at the highest Agolin inclusion rate (6%) compared to the lower Agolin inclusion rates (3% and 4.5%) but did not differ to that when no Agolin was included or when Agolin or uPRO were included alone. The relative amount of CH₄ within total gas produced was lowest (P < 0.001) at the highest Agolin inclusion rate. The digestibility of the Rhodes grass hay and the molar proportion of VFA were unaffected (P = 0.735 and P = 0.308, respectively) by the uPRO and Agolin treatments.

Table 3.

Apparent IVDMD, apparent IVOMD, total gas production, methane production, ratio of total gas to CH₄ production, and VFA of Rhodes grass hay incubated with Agolin Ruminant L (Agolin; Alltech Technology) and uPRO ORANGE (uPRO; DIT Ag-Tech, QLD, Australia) simulated water-based supplementation (mean ± SEM)

Variable	Treatment1						P-value2
	Rhodes control	Agolin	uPRO	3% Agolin + uPRO	4.5% Agolin + uPRO	6% Agolin + uPRO
Apparent IVDMD, %	42.7 ± 1.52	41.9 ± 0.76	42.0 ± 0.48	40.7 ± 0.86	43.5 ± 1.78	42.8 ± 1.64	0.735
Apparent IVOMD, %	42.5 ± 1.45	40.5 ± 0.90	41.9 ± 0.45	40.7 ± 0.86	43.6 ± 1.64	42.7 ± 1.61	0.495
Total gas production, mL/g DM	59.9 ± 2.40b	60.4 ± 1.12b	55.8 ± 1.69b	67.7 ± 1.67ab	75.0 ± 0.27a	66.8 ± 6.00ab	0.006
CH4 production, mL/g DM	6.4 ± 0.36ab	6.3 ± 0.07ab	6.3 ± 0.20ab	7.6 ± 0.25a	7.3 ± 0.02a	5.4 ± 0.55b	0.003
Total gas: CH4	10.8 ± 0.31ab	10.5 ± 0.14ab	11.3 ± 0.03a	11.2 ± 0.10a	9.8 ± 0.05b	8.1 ± 0.45c	<0.001
Total VFA, mM	7.8 ± 0.51	7.8 ± 0.43	7.5 ± 0.80	5.4 ± 0.65	7.5 ± 0.61	8.4 ± 1.61	0.308
Acetic, mM	4.3 ± 0.24	4.4 ± 0.21	4.0 ± 0.52	3.2 ± 0.34	4.6 ± 0.35	5.0 ± 1.02	0.290
Propionic, mM	1.5 ± 0.12	1.4 ± 0.04	1.3 ± 0.09	1.1 ± 0.11	1.6 ± 0.07	1.6 ± 0.25	0.134
Butyric, mM	0.5 ± 0.03	0.5 ± 0.02	0.5 ± 0.03	0.4 ± 0.04	0.5 ± 0.06	0.5 ± 0.05	0.239
Valeric, mM	0.7 ± 0.04	0.7 ± 0.05	0.8 ± 0.02	0.6 ± 0.05	0.6 ± 0.05	0.7 ± 0.05	0.156
Total branched chain VFA, mM	0.7 ± 0.10	0.8 ± 0.13	0.9 ± 0.15	0.1 ± 0.18	0.3 ± 0.16	0.5 ± 0.24	0.078
Acetic:propionic	3.0 ± 0.11	3.2 ± 0.06	3.1 ± 0.21	2.9 ± 0.08	2.9 ± 0.14	3.1 ± 0.18	0.671

¹Treatments: Agolin is 50 mg Agolin Ruminant L/L water, uPRO is 1.7 mL uPRO/L water, 3% Agolin + uPRO is 3% Agolin Ruminant L/1.7 mL uPRO/L water, 4.5% Agolin + uPRO is 4.5% Agolin Ruminant L/1.7 mL uPRO/L water, and 6% Agolin + uPRO is 6% Agolin Ruminant L/1.7 mL uPRO/L water.

²Entries in the same row with different superscript letters indicate significant differences where the threshold for significance is P < 0.05.

Experiment 2

Steers consuming water with uPRO and the lower Agolin inclusion rate (6 µL/L; T2) had lower (P < 0.0001) daily CH₄ emissions (g CH₄/d) than steers consuming water with uPRO alone (T1) or uPRO and the greater Agolin inclusion rate (T3; Table 4). These differences were particularly pronounced during weeks 4, 6, 7, and 8 (Figure 2). Methane yield (g CH₄/kg DMI) and CH₄ intensity (g CH₄/kg LWG) from steers were unaffected (P = 0.11 and P = 0.20, respectively) by the inclusion of Agolin in the drinking water containing uPRO. Carbon dioxide emissions were greater (P = 0.001) from steers consuming water containing Agolin at the greater inclusion rate (T3) compared to steers consuming the other treatment (T2) or the control (T1). Methane yield (g CH₄/kg DMI) and CH₄ ratio (g CH₄/L water intake) declined over the experimental period and were lower in weeks 4 to 8 and weeks 6 to 8, respectively, compared to week 1 (Figure 2). There were significant treatment × time interactions, where CH₄ emissions were lower from steers consuming the low dose of Agolin (T2) in the drinking water compared to those consuming the high dose of Agolin (T3) in weeks 7 and 8, with steers consuming the low dose of Agolin (T2) in the drinking water emitting 17.6% (P = 0.0205) less CH₄ than steers consuming water without Agolin (T1) in week 8 (Figure 2). In addition, daily CH₄ emissions of steers consuming the low dose of Agolin declined by 16.4% from week 1 to week 8 of the experimental period (P = 0.0027). Methane yield was lower for steers consuming the T2 compared to T3 in week 8, but there were no other differences between treatments across the experimental period.

Table 4.

Mean emissions of CH₄, H₂, and carbon dioxide CO₂, and CH₄ yield, CH₄:water intake ratio, GEM unit visits, LW, LWG, NDFD, and feed and water intake of steers consuming water containing uPRO ORANGE (DIT Ag-Tech, QLD, Australia) alone or with 2 inclusion rates of Agolin Ruminant L (Alltech Technology)

Parameter	Treatment1			SEM	P-value2
	T1	T2	T3		Treat	Time	Treat × Time3
CH4, g/d	138a	131b	143a	0.72	<0.0001	<0.0001	<0.0001
Hydrogen, g/d	0.61	0.62	0.67	0.01	0.05	<0.0001	0.03
Carbon dioxide, g/d	5,435b	5,361b	5,650a	22.66	0.001	<0.0001	0.27
CH4 yield, g/kg DMI	22.59	22.02	23.05	0.22	0.11	<0.0001	0.002
CH4 intensity, g/kg LWG4	296.66	236.62	251.23	13.79	0.20	–	–
CH4:water intake, g/L	9.29ab	7.89b	9.17a	0.13	0.002	<0.0001	<0.0001
GEM visits, events/steer/week	24.0	24.4	25.5	0.71	0.92	<0.0001	0.58
LW, kg	281	280	284	1.86	0.95	<0.0001	0.21
LWG, kg/d	0.49	0.56	0.60	0.06	0.33	0.001	0.21
Intake
Hay, kg/d	5.98	5.65	5.89	0.06	0.19	<0.0001	0.05
Pellets, kg/d	0.56	0.55	0.58	0.10	0.85	<0.0001	0.96
Total feed, kg/d	6.52	6.16	6.44	0.06	0.09	<0.0001	0.02
Water, L/d	17.60	17.30	17.50	0.17	0.90	<0.0001	0.02
NDFD, %	60.20	59.10	59.90	0.48	0.64	–	–

¹Treatments (Treat): T1 is a mix of fresh water + uPRO ORANGE, T2 is a mix of fresh water + uPRO ORANGE + Agolin Ruminant L (6 µL/L), and T3 is a mix of fresh water + uPRO ORANGE + Agolin Ruminant L(24 µL/L).

²Entries in the same row with different superscript letters indicate significant differences where the threshold for significance is P < 0.05.

³Treat × Time is the interaction between treatment and measurement periods.

⁴Average of CH₄ production per steer compared to the average of weekly LWG per steer across an entire experimental period.

Figure 2.

The change in (A) CH₄ emissions, (B) hydrogen emissions, (C) CH₄ yield, and (D) CH₄ ratio relative to water intake of steers consuming water containing uPRO ORANGE (uPRO; DIT Ag-Tech, QLD, Australia) alone (treatment 1) or water and uPRO with low (treatment 2) and high (treatment 3) inclusion rates of Agolin Ruminant L (Alltech Technology). Asterisk indicates the location of significant differences where P < 0.05.

There were no differences in LW (P = 0.95), LWG (P = 0.33), feed intake (P = 0.09), water intake (P = 0.9), or NDFD (P = 0.64) for steers consuming the different treatments (Table 4). Similarly, the concentration of all analytes measured in the plasma and all rumen fermentation parameters of steers were unaffected by the inclusion of Agolin in the drinking water containing uPRO (Table 5).

Table 5.

The pH and concentration of VFA in the rumen fluid and the concentration of analytes in the plasma of steers consuming water containing uPRO ORANGE (DIT Ag-Tech, QLD, Australia) alone or with 2 inclusion rates of Agolin Ruminant L (Alltech Technology; mean ± SEM)

Parameters	Treatment1			P-value
	T1	T2	T3
Rumen
Total VFA, mM	80.4 ± 2.88	80.3 ± 3.27	79.0 ± 3.64	0.941
Acetic, mM	56.7 ± 2.23	55.8 ± 1.70	55.3 ± 2.56	0.907
Propionic, mM	10.1 ± 0.44	10.1 ± 0.44	10.0 ± 0.43	0.967
Butyric, mM	4.2 ± 0.20	4.2 ± 0.34	4.3 ± 0.20	0.948
Valeric, mM	3.4 ± 0.18	3.7 ± 0.44	3.3 ± 0.18	0.619
Hexanoic, mM	0.2 ± 0.01	0.2 ± 0.02	0.1 ± 0.01	0.577
Total bcVFA2, mM	5.8 ± 0.34	6.3 ± 0.59	5.9 ± 0.37	0.717
Acetic:propionic	5.6 ± 0.08	5.6 ± 0.15	5.5 ± 0.04	0.927
pH	6.87 ± 0.08	6.88 ± 0.09	6.85 ± 0.09	0.983
Plasma
PUN3, mmol/L	2.9 ± 0.16	2.4 ± 0.15	2.9 ± 0.15	0.078
Phosphate, mmol/L	2.4 ± 0.06	2.4 ± 0.06	2.5 ± 0.06	0.316
Glucose, mmol/L	4.3 ± 0.12	4.4 ± 0.11	4.4 ± 0.11	0.797

¹Treatments are, T1 is a mix of fresh water + uPRO ORANGE, T2 is a mix of fresh water + uPRO ORANGE + Agolin Ruminant L (6 µL/L), and T3 is a mix of fresh water + uPRO ORANGE + Agolin Ruminant L (24 µL/L).

²Total bcVFA is total branched-chain VFA (isobutyric, isovaleric).

³PUN is plasma urea nitrogen.

Discussion

Methane emissions

Both the in vivo and in vitro experiments demonstrated the delivery of Agolin via the water-reduced CH₄ production with no effect on indicators of rumen fermentation (experiments 1 and 2) or steer productivity (experiment 2). However, the results of experiment 1, particularly those related to CH₄ production, do not directly align with those of experiment 2. Quantities of Agolin at ~50 mg/L (ca. 700 mg/d) were ineffective in the short term in vitro fermentation, whilst a much lower delivery rate of ~5 mg/L (ca. 100 mg/d) resulted in a significant reduction in CH₄ emissions in vivo. Similar observations were reported by Castro-Montoya et al. (2015), who observed CH₄ reductions relative to body weight in Belgian Blue beef heifers fed maize silage supplemented with Agolin at a rate of 200 mg/d but reported no reduction in CH₄ production in 24 h in vitro fermentations using aqueous solutions of Agolin (150 and 300 mg/L). The extended period required to observe statistically significant CH₄ reductions is attributed to the action of essential oils as rumen modifiers, which take time to affect rumen microbial populations (Calsamiglia et al., 2007). This delayed response has been consistently observed in the in vivo experiments using Agolin, where modulation of the rumen microbial population takes approximately 4 wk (Belanche et al., 2020). The increases in total gas associated with lower levels of Agolin supplementation in experiment 1 are also of interest, but require data on H₂ and CO₂ for appropriate interpretation, and capturing this information in future research may be of value. The results of experiment 1 establish the concentration of Agolin required for a significant reduction in CH₄ emissions, but they are so high that the usefulness of further in vitro analysis when investigating water-delivered essential oil-based compounds may be limited. In general, in vitro incubations are useful for ranking treatments based on their relative effects, but they do not provide precise values for actual in vivo outcomes. Their role is more to establish trends and compare treatments rather than to predict the responses under practical conditions.

This study found no significant decrease in CH₄ yield (g CH₄/kg DMI) consistent with other studies with beef cattle (Castro-Montoya et al., 2015). Castro-Montoya et al. (2015) attributed this to a comparatively greater LWG in the cattle consuming Agolin. In contrast, a high dose of Agolin (ca. 183 mg/kg feed offered) did not affect on CH₄ production of dairy-beef Holstein Friesian steers, but a significant decrease in CH₄ yield was reported (Miller et al., 2023). This decrease in CH₄ yield was attributed to a lower feed intake in control steers when they were isolated in respiration chambers, making a comparison to the data in this study difficult. Both these findings contrast with those of experiment 2, where Agolin had no significant effect on LWG or feed intake. No significant impact was observed on CH₄ emissions for the high Agolin inclusion rate used in experiment 2. There may be a correlation with high dose rates of Agolin being less effective in terms of CH₄ mitigation over short time periods. In a 56 d study on Holstein dairy, cows fed Agolin at a rate of 1 g/animal/d, Carrazco et al. (2020) observed no differences in CH₄ production or yield. Conversely, Bach et al. (2023) observed decreases in CH₄ production and yield from Holstein cows when feeding Agolin at 1 g/animal/d over a 91 d study.

Despite the positive effect of the lower concentration of Agolin on CH₄ emissions from cattle, this result potentially highlights that there may be some unique challenges faced with the water delivery of Agolin. It has been established that up to 20% of water consumed, and hence any bioactive compound contained therein, can directly bypass the rumen via the closing of the esophageal groove, directly entering the abomasum (Woodford et al., 1984; Panjaitan et al., 2010). Despite being thought to only be active in calves, the receptors in the back of the mouth responsible for this reflex can be triggered in adults as a response to drugs such as metoclopramide and vasopressin, or high concentrations of mineral salts or other water contaminants (Carruthers et al., 1994; Ivany et al., 2002; Wagner and Engle, 2021). There is the potential that the lack of a significant CH₄ reduction using the greater concentration of Agolin in experiment 2 could be a result of the triggering of this mechanism, and therefore an inability of the compound to influence the rumen microbiome. If this is the case, the dosage rates of 3, 4.5, and 6 % Agolin used in experiment 1 would also not be feasible for in vivo applications.

Methane measurements

The method used to measure CH₄ in this trial has been well-established for some time. Although the GEM system is straightforward to implement, certain limitations require caution in its use. Hammond et al. (2016) recommended spreading GEM measurements over 24 h or adjusting them to avoid bias from clustered visits and to account for cattle’s natural circadian CH₄ emission patterns. Another consideration is the daily variation of CH₄ emissions. The findings of Manafiazar et al. (2016) suggest that averaging over 7 to 14 d with at least 20 spot samples is essential for reliable CH₄ and CO₂ emissions. In this study, the GEM unit was set to allow up to 5 feeding events per animal per day, with a minimum interval of 4 h between visits, and used average values over 7 d. The number of visits per animal varied weekly, averaging 25 visits with an SD of 9. In the current work, the analysis was conducted using daily averages to reduce potential bias because animals with fewer visits were not weighted equally with those visiting more frequently. There was no significant difference in the number of visits between treatments or treatment interactions over time. A significant time effect was observed, with fewer visits in the first and last weeks (17 to 20 visits) compared to the middle weeks (27 to 30 visits). This likely reflects animals adjusting to the equipment early on and the interference of daily fecal collections during the final week. The use of at least 30 visits of 2 min duration to calculate CH₄ production has been suggested (Arthur et al., 2017; Beck et al., 2024). This was achieved in most weeks in the current work. Despite the lower number of visits in the first and final weeks, the analysis of daily CH₄ emissions showed no difference between the middle and final weeks when considering all treatments together, indicating that the number of visits did not affect emission results. In addition, no treatment had a lower number of visits in any specific week, ensuring no impact on the analysis. Regarding the duration of each visit, the study of Authur et al. (2017) found that using GEM records with a minimum of 2 min reduced precision compared to a minimum of 3 min. In their work, more records were needed to achieve the same precision with 2 min visits and the authors concluded that to accurately compute CH₄ or CO₂ production rates, at least 30 records with a minimum GEM visit duration of 3 min are necessary. In the current work, the analysis included visits with a minimum duration of 2 min. However, the average visit duration averaged 3:34 min.

Another point to consider is the circulation of air, as airflow rates through the GEM impact the quality of CO₂ and CH₄ emission estimates (Gunter and Beck, 2018), with incomplete breath capture hypothesized at lower airflow rates. Gunter et al. (2017) evaluated 758 estimates, with airflow ranging from 10.7 to 36.6 L/s, and found that emission estimates were unaffected when airflow was between 26.0 and 36.6 L/s. However, airflow rates below 26.0 L/s resulted in significantly lower estimates for both gases, likely due to incomplete capture of the breath cloud. Therefore, it was concluded that maintaining airflow rates above 26 L/s is crucial for accurate emission measurements. In the current study, airflow rates averaged 44.7 L/s, with the lowest value used of 40.34 L/s.

Dry matter intake

Muetzel et al. (2024) investigated the effects of animal and dietary parameters on CH₄ emissions in cattle and concluded that DMI is the primary driver of emissions, with diet quality having minimal impact when DMI is similar. Beauchemin and McGinn (2006) examined the potential of grain supplementation to reduce CH₄ emissions in growing feedlot cattle. Their findings indicated that altering forage-to-concentrate ratios and grain sources did not significantly affect CH₄ emissions for animals with the same DMI. In the present study, there were no significant effects on DMI between treatments during any of the experimental weeks. Although a trend was noted, as indicated by a P-value of 0.09 for total DMI, caution is advised in the interpretation, particularly given that the average weekly DMI analysis revealed that the primary contributor to the trending P-value was data from week 1 of the experiment, during which no effects of Agolin on rumen fermentation or CH₄ production were anticipated.

Productivity and rumen fermentation

No significant differences in productivity (experiment 2) or rumen fermentation parameters (experiments 1 and 2) were observed in these experiments. The lack of any significant impact on DMI by steers in response to Agolin treatment in experiment 2 is consistent with that reported elsewhere (Belanche et al., 2020). As mentioned previously, almost all the studies investigating Agolin are based on the productivity metrics of dairy cattle, with most not reporting LW, LWG, or feed conversion efficiency. However, Miller et al. (2023) reported no differences in LWG, DMI, or LW in their study on dairy-beef Holstein Friesian steers, and Hart et al. (2019) observed no differences in the LW of Holstein Friesian dairy cows at the end of their study. These observations are consistent with the results of the current study. Additionally, there were no effects of Agolin on water intake, regardless of the concentration of Agolin included in the water. This was expected and is in line with other studies on the impact of water intake utilizing supplementation where this was also the case (McLennan et al., 1991; Romanzini et al., 2024).

Regarding rumen fermentation, separate review articles on the use of essential oils as rumen modifiers for CH₄ mitigation suggest that ideally, these treatments should result in increased propionate and butyrate concentrations, as well as a reduction in the acetate to propionate ratio and ammonia N (Calsamiglia et al., 2007; Cobellis et al., 2016). This stems from the inhibition of methanogenic and hyperammonia-producing microbial species and increased availability of hydrogen for propionate synthesizing microbes (Króliczewska et al., 2023). However, Belanche et al. (2020) note in their meta-analysis that in most of the studies they reviewed, Agolin is unlikely to change VFA concentration or proportions, an observation also made in both experiments 1 and 2 in the current study. It appears that much of the VFA data from the Belanche et al. (2020) paper have not yet been reported, and only one paper available reports no changes to molar proportions of VFAs in dairy cows except for lower proportions of propionic acid (Elcoso et al., 2019). More recently, similar results for VFA concentrations to those reported in this study were reported by Silvestre et al. (2023); however, in their analysis of Holstein dairy cows, they only observed a reduction in CH₄ intensity when compared to milk production parameters, but not in overall CH₄ production or yield in terms of DMI.

The significantly increased CO₂ production observed in the T3 treatment could be associated with an increase in fermentation, however, this did not translate into significant changes in VFA concentrations or LWG. The studies of Silvestre et al. (2023) and Carrazco et al. (2020) reported no significant differences in CO₂ production when supplementing dairy cattle with Agolin at 1 g/animal/d in mixed rations. The increased CO₂ production observed when aqueously delivering Agolin at 400 mg/animal/d in the current study may be due to diet and dose specificity, considering the different interactions of the product with digesta when administered through water. The increase in CO₂ when using additives to reduce CH₄ emissions has been previously reported (Glasson et al., 2022), raising questions about the potential consequences of altering the natural balance of the rumen environment. Pursuing CH₄ reductions without fully understanding the broader impacts to animals, consumers, and the environment may lead to unintended negative outcomes. Therefore, a greater understanding and balanced approach is essential for achieving sustainable solutions.

Regard must also be given to the low pH of the water supplied to steers in this study, resulting from the addition of a supplement containing urea phosphate. Urea phosphate, a compound formed by the reaction of urea and phosphoric acid, is known for its acidifying effect. When dissolved in water, urea phosphate dissociates, releasing hydrogen ions (H⁺), which leads to a decrease in pH. This property makes it useful in agricultural applications, particularly for lowering the pH of alkaline soils and improving nutrient availability for plants (Hussein et al., 2022). However, water supplementation of urea phosphate and the effect that lowering water pH may have on rumen fermentation is not well understood. Ruminal pH values of 5.6 and 5.2 are often considered thresholds for chronic and acute acidosis, respectively (Owens et al., 1998). Despite the low pH of water imbibed, the average measured pH of rumen fluid samples from individual steers in this study was above 6.8 and exceeded those reported (between 6.1 and 6.8) by Bowen et al. (2016) in beef steers grazing a variety of tropical pastures. Owens et al. (1998) highlight a key factor in feedlot diets is that most lactate-utilizing microbes are sensitive to low pH. However, the basal diet in this trial was low in starch, making the rumen unlikely to have a high presence of lactate-utilizing microbes. Another factor to consider is the behavior of animals ingesting water. Steers in our experiment did not consume water in one single event. However, this may be the case for grazing animals, so considerations would need to be made for extensive applications.

Costa et al. (2019) reported that cattle grazing wet-season tropical grasses had a ruminal liquid volume ranging from 42 to 61 L, or about 5% to 8% of their body weight. In this trial, the steers weighed approximately 250 kg, which would imply that they had more than 10 L of liquid in their rumen if similar proportions were applied. With an average water intake of 14.5 L over a 24 h period, even if an animal consumed all the water in one instance (which was not the case), the low pH water would be diluted with a similar volume of ruminal fluid containing buffering agents. These factors combined suggest that the low pH of the water did not negatively impact rumen activity.

Limitations and future research

Whilst the results reported in this study provide compelling evidence for the efficacy of Agolin delivered through the drinking water, there is a large scope for further investigation to determine the precise mechanism for CH₄ reduction and the extent of its effectiveness. For example, bioactive compounds introduced to the rumen in liquid form may mix and interact with digesta at different rates and retention times compared to those delivered in solid-form fed compounds.

At the behavioral level, asynchrony of the timing of eating (grazing) and drinking events witnessed under extensive conditions may also influence the bioactive potential of compounds entering the rumen and the fermentation of feed substrates in the rumen. Whether there is an optimum time to deliver compounds to the rumen in relation to fermentation times after eating is unknown at this stage. It is also noted that both the in vitro and in vivo experiments only tested responses to a single basal feed (medium-quality Rhodes grass hay), and that the capacity to shift fermentation characteristics are often constrained to upper and lower thresholds by the quality of the basal feed. The medium-quality Rhodes grass hay, which is representative of many pastures found in extensive grazing systems in subtropical and tropical regions, would have lower fermentation characteristics than a high grain ration. The results obtained from NDFD analysis support this, and although the technique has been shown to produce accurate measurements, the authors acknowledge the limitations in performing grab samples at only one time point, instead of total fecal collections on at least one animal for confirmation (Velasquez et al., 2018). To address this limitation, the collections were repeated over a 5-day period to better represent the various interactions. This procedure was chosen due to challenges in obtaining total fecal output in the group pen experimental design and a desire to reduce animal handling whilst using the GEM. However, the feed was not deficient in any nutrients (N or metabolizable energy) to meet the productivity targets for this class of cattle, so the capacity to shift rumen fermentation characteristics as one might expect with a nutrient-deficient basal diet was not possible.

Furthermore, greater CH₄ production and yield reductions may also result from longer periods of supplementation. Miller et al. (2023) reported a greater statistical difference in CH₄ yield after 46 d of treatment (P < 0.05) as opposed to 116 d (P < 0.01) when compared to the control, and though this was attributed to DMI reduction in control animals, it is a trend that is supported by the steady decrease in CH₄ production associated with the low Agolin concentration treatment in experiment 2 (Figure 2). Additionally, although not significant in the current study, the variations in LWG between control and low Agolin dosage (0.49 vs. 0.56 kg/d) observed may further contribute to potential CH₄ reductions because of shorter growing periods to reach market LW targets. Further longer-term studies would also be useful to investigate the potential for microbial adaptation, which is believed to be responsible for some of the inconsistent and revertive CH₄ production results observed in some studies, or that may result from supplementation of bioactive compounds at greater concentrations (Cobellis et al., 2016; Belanche et al., 2020). Studies that incorporate detailed analysis and profiling of the rumen microbiome at various stages may be particularly valuable in this regard.

Conclusions

This is one of the first studies investigating the efficacy of Agolin to reduce CH₄ emissions from beef cattle when delivered through drinking water. Both in vitro and in vivo experiments were conducted using Agolin dissolved in water at various concentrations. Short-term in vitro analysis revealed no reduction in CH₄ production at low Agolin inclusion rates, consistent with the literature, and required inclusion rates of Agolin above recommended levels to reduce CH₄ production, suggesting limited efficacy of in vitro analysis of water-based Agolin solutions. In contrast, the lower rate of Agolin inclusion delivered through the drinking water to steers resulted in a total reduction in CH₄ emissions of 16.4% over 56 d from steers consuming a medium-quality roughage diet ad libitum. It appears that Agolin provided via the drinking water can reduce CH₄ emissions, but the modulation of the rumen environment that is necessary for this to occur takes some time. Further research is required to determine long-term effects, the optimal delivery rate, the mode of action, and any rumen bypass, the efficacy when delivered via drinking water to cattle fed other diets, and when provided in water without any other supplements or additives.

Glossary

Abbreviations

CP

crude protein

DM

dry matter

DMI

dry matter intake

GEM

GreenFeed emissions monitoring unit

IVDMD

in vitro dry matter digestibility

IVOMD

in vitro organic matter digestibility

LW

live weight

LWG

live-weight gain

NDF

neutral detergent fiber

NDFD

neutral detergent fiber digestibility

NPN

nonprotein nitrogen

OM

organic matter

PUN

plasma urea nitrogen

RFID

radio-frequency identification

VFA

volatile fatty acid

Conflict of interest statement

The authors have the following competing interest: the second author (E.P.R.) is an employee of DIT Ag-Tech that provided the dosing technology and uPRO supplements. The other authors have no conflicts of interest.

Author Contributions

Ryan Batley (Conceptualization, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review & editing), Elieder Romanzini (Data curation, Formal analysis, Investigation, Writing—review & editing), Kawane D. da Silva (Investigation), William L. de Souza (Investigation, Methodology, Writing—review & editing), Simon P. Quigley (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing—review & editing), Karen J. Harper (Validation, Writing—review & editing), Mark Trotter (Funding acquisition, Supervision, Writing—review & editing), Priscila A. Bernardes (Data curation, Formal analysis, Software, Writing—review & editing), Mani Naiker (Conceptualization, Supervision, Writing—review & editing), and Diogo F.A. Costa (Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing—review & editing)