The Mitigation of Methane Emissions from Ruminants: Evaluating the Efficacy of Selected Additives and Feed Replacements in an In Vitro Trial

Abstract

The need for new feed ingredients that could reduce methane (CH₄) emissions from dairy cattle while maintaining rumen function is essential for sustainable milk production. This study aimed to evaluate the CH₄ mitigation potential of selected microalgae and macroalgae, along with an agro-industrial by-product, using two feeding strategies, and hypothesized that lipid- and polyphenol-rich materials would reduce CH₄ production in an inclusion-dependent manner. An in vitro batch culture study (24 h) was conducted to evaluate microalgae (Euglena gracilis and Aurantiochytrium spp.), macroalgae (Undaria pinnatifida), and an agro-industrial by-product (grape marc) either as feed additives (5%) or as a partial replacement of the concentrate mixture (30%, 50%, and 70%) in a basal diet consisting of 50% Klein grass hay and 50% concentrate mixture. As a feed additive, grape marc stands out for its potential to reduce CH₄ yield by about 43.3% without adversely affecting digestibility, pH, or total volatile fatty acid concentrations. When used as feed replacements, Euglena-, Aurantiochytrium-, and grape marc-based feeds reduced CH₄ yield at the highest replacement levels (50 and 70%); however, these effects were accompanied by decreased total gas production and volatile fatty acid concentrations, indicating reduced fermentation activity. Meanwhile, at a 30% replacement level, they showed promising efficiency as alternative feeds. Overall, CH₄ mitigation depends more strongly on inclusion strategy rather than feed type. Lipid-rich microalgae showed potential as concentrate replacements up to 30%, whereas grape marc was most effective as a feed additive for reducing CH₄ emissions.

1. Introduction

Livestock production plays a crucial role in global food systems by providing high-quality protein and essential nutrients, with ruminants, particularly dairy cattle, being a significant contributor to this supply [1]. However, ruminant production is associated with substantial environmental challenges, particularly methane (CH₄) emissions from enteric fermentation [2,3]. The CH₄ not only contributes to climate change but also represents an energy loss of approximately 2–12% of gross dietary intake, thereby reducing feed efficiency and animal productivity [4]. Therefore, the development of effective nutritional strategies to mitigate enteric CH₄ emissions while maintaining normal rumen function is essential for improving the environmental sustainability of dairy production systems.

Various mitigation strategies have been explored to address enteric CH₄ emissions, including dietary manipulation, improved management practices, selective breeding, vaccine-based approaches, and direct manipulation of the rumen microbiome [3,5]. Among these approaches, feed-based strategies are particularly attractive due to their practicality and direct influence on rumen fermentation. Dietary supplementation with lipids, tannins, essential oils, macroalgae, and other bioactive compounds has shown potential to suppress CH₄ production and improve rumen fermentation efficiency [6,7]. Lipid-rich feeds can reduce CH₄ production by decreasing hydrogen (H₂) availability and inhibiting methanogenic activity [8], while polyphenolic compounds, such as tannins and phlorotannins, may suppress methanogenesis through direct antimicrobial effects and by altering microbial fermentation pathways [3,5,7]. However, these compounds may also influence nutrient digestibility and overall fermentation efficiency depending on their inclusion level and mode of application. As a result, nutritional approaches that affect rumen microbial ecology and fermentation kinetics have emerged as an effective strategy for mitigating CH₄ production in dairy cattle [5,9].

Building on this premise, the current study examined how various micro- and macroalgae products and agro-industrial feed by-products affect CH₄ production, nutrient digestibility, and rumen fermentation. The novel products used in this study are readily available in Japan and include strains of Euglena gracilis, a microalga grown under photoautotrophic (green) and heterotrophic (yellow) conditions; three commercial Aurantiochytrium-based products (A, C, and D); brown seaweed (Undaria pinnatifida); and grape marc. Previous studies suggest that green Euglena has the potential to reduce CH₄ due to its high fat content (10–20%) and fatty acid composition [10,11]. Whereas the yellow strain, despite its higher lipid content (24.81 wt% on a dry weight basis) [12], has not been evaluated in this context. Aurantiochytrium spp. are the heterotrophic marine protists that are frequently referred to as microalgae in applied research due to their algal-like cultivation and high production of polyunsaturated fatty acid (PUFA), particularly docosahexaenoic acid (DHA) [13,14]. Aurantiochytrium sp. SD116 has been reported to accumulate up to 56.3% lipid on a dry cell weight basis, with docosahexaenoic acid (DHA) accounting for approximately 50% of total fatty acids [15]. In the present study, three commercial Aurantiochytrium-based products (A, C, and D), differing in lipid content and antioxidant functionality according to the manufacturer, were used, rather than taxonomically defined strains. Although Aurantiochytrium has not been directly evaluated for rumen fermentation or CH₄ mitigation, dietary supplementation with Aurantiochytrium limacinum has been reported to show a tendency to improve productivity and milk composition in lactating dairy cows without adversely affecting health status [16]. Undaria pinnatifida is a brown seaweed rich in polyphenolic compounds, particularly phlorotannins [17], as well as fermentable polysaccharides [18], both of which are known to influence rumen fermentation and microbial communities. Previous studies have demonstrated that supplementation with U. pinnatifida extract, used as a bioactive additive, reduces CH₄ production in vitro without impairing overall rumen fermentation, an effect attributed primarily to polyphenolic modulation of rumen microbial populations [19]. Grape marc, a by-product of winemaking, and rich in tannins and lipids, has been found to decrease CH₄ emissions in dairy cattle by up to 20% [20], although the tannins showed variable impacts on fiber digestibility and rumen microorganism activity [21].

Although various feed resources, including micro- and macroalgae, and agro-industrial by-products, have shown the potential to mitigate CH₄ emissions, most studies have evaluated these materials either as feed additives or as dietary ingredients, but rarely compared both strategies within the same experimental framework. Furthermore, information is limited regarding how the inclusion level and mode of application influence the balance between CH₄ mitigation, rumen fermentation, and nutrient digestibility. Therefore, this study aimed to evaluate the CH₄ mitigation potential of selected micro- and macroalgae products and agro-industrial by-products using two practical feeding strategies: (1) inclusion as a feed additive (5%), and (2) partial replacement of the concentrate mixture at increasing levels. This dual approach was designed to distinguish the mode of action of each material and to determine the optimal method of its application in ruminant diets. It was hypothesized that lipid-rich and polyphenol-containing feed resources would reduce CH₄ production by limiting H₂ availability and inhibiting methanogenic activity, while their effects on rumen fermentation and nutrient digestibility would depend on the inclusion level and feeding strategy.

2. Materials and Methods

The research was carried out at Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Hokkaido, Japan. All procedures were conducted following the ethical standards established by the university’s Animal Care and Use Committee (approval number 24-79). The animals involved in the research were housed and cared for by the Field Science Center at Obihiro University.

2.1. Basal Diet and Tested Materials

The basal diet in this research consisted of finely chopped Klein grass (Panicum coloratum) hay, with particles approximately 1 mm in size, and commercial concentrate mixture (in a ground form with particle size of about 1 mm) from Alpha-Kotan, Chubu Shiryo Co., Ltd., Nagoya, Japan. The tested materials in this study included two types of Euglena gracilis, E. gracilis yellow (EGY) and E. gracilis green (EGG); three commercial Aurantiochytrium-based products A, C, and D (AA, AC, and AD, respectively); Undaria pinnatifida (UP); and grape marc (GM). Euglena types were supplied in fine powder form (particle size range: 100–150 µm), 100% pure, with no excipients. Similarly, three commercial Aurantiochytrium-based products (A, C, and D) were supplied in powder form (100 µm), 100% pure with no excipients. According to the manufacturer, product AA contained lower lipid levels, product AC contained higher lipid levels, and product AD contained higher lipid levels with additional antioxidants. The GM, a by-product of wine making made up of grape skins, seeds, stalks, and stems, was provided in a ground form with an approximate particle size of 1 mm. Brown seaweed (UP) was supplied as dried whole material and subsequently ground using a cutting mill (Retsch SM 2000, Retsch Gmbn, Haan, Germany) to pass through a 1 mm sieve. All tested materials were supplied by Euglena Co., Ltd., Tokyo, Japan. The chemical compositions of the materials used in this study are presented in

Table 1. Chemical composition (g/kg in dry matter) of the feeds used in this study.

g/kg	KG 1	CM 2	EGY 3	EGG 4	AA 5	AC 6	AD 7	UP 8	GM 9
Dry matter	952	867	924	960	917	966	969	949	951
Organic matter	906	941	962	964	945	911	912	551	909
Crude ash	94	59	38	36	55	89	88	449	91
Crude protein	116	205	116	254	217	129	134	102	114
Ether extracts	21	32	32	101	72	234	224	31	81
Neutral detergent fiber	648	361	301	125	316	249	237	343	453
Acid detergent fiber	373	94	54	82	222	129	122	185	335
Acid detergent lignin	50	19	0	0	0	0	0	8	197
Non-fiber carbohydrate	121	343	513	484	340	229	317	117	261

¹ KG: Klein grass. ² CM: concentrate mixture. ³ EGY: Euglena gracilis yellow. ⁴ EGG: Euglena gracilis green. ⁵ AA: Aurantiochytrium A. ⁶ AC: Aurantiochytrium C. ⁷ AD: Aurantiochytrium D. ⁸ UP: Undaria pinnatifida. ⁹ GM: grape marc.

.

2.2. Rumen Fluid Collection

Rumen fluid was collected from two ruminally cannulated non-lactating Holstein cows (approximately 10 years old) with an average body weight of 894 kg. These cows were maintained on a diet consisting of orchard grass (Dactylis glomerata) hay at a maintenance level based on the Japanese feeding standard for dairy cattle [22], with free access to clean water, and a mineral block (Koen SELENICS TZ, Nippon Zenyaku Kogyo Co., Koriyama, Japan). The diet contained organic matter (OM) at 920 g/kg, crude protein (CP) at 99 g/kg, ether extracts (EE) at 24 g/kg, neutral detergent fiber (NDF) at 711 g/kg, acid detergent fiber (ADF) at 433 g/kg, and acid detergent lignin (ADL) at 46 g/kg. Rumen fluid was collected at 10:00 AM, approximately 2 h after feeding. About 3 L of rumen fluid was collected from four distinct locations within the rumen of each cow. The collected rumen fluid from each cow was then filtered through a four-layer gauze, and sequentially combined in approximately equal volumes in a thermos flask that had been pre-warmed at 39 °C. During transportation, the thermos flask was placed in an insulated container with warmed water (37–39 °C) to maintain microbial viability and immediately transported to the laboratory within 20 min.

2.3. Experimental Design

This research utilized two different experimental designs (EXP 1 and EXP 2) to represent two feeding strategies: as a feed additive and partial concentrate replacement. The feed additive (5%) in EXP 1 was selected to evaluate the direct additive effects of the tested materials. In contrast, EXP 2 was designed to assess the effects of partial replacement of the concentrate mixture over a wide range of inclusion levels (30%, 50%, and 70%) to examine dose-dependent responses and the practical feasibility of using these materials as feed ingredients. The first design (EXP 1) included a control group that was provided with a basal diet consisting of 500 mg of fresh matter, made up of 50% Klein grass hay and 50% concentrate mixture. The feed additives (EGY, EGG, AA, AC, AD, UP, and GM) were incorporated directly as additives at a rate of 5% of the substrate. The second experimental design (EXP 2) utilized the same control diet as EXP 1; however, the feed replacements (EGY, EGG, AA, AC, AD, UP, and GM) were mixed with the basal diet to partially substitute the concentrate mixture. Each material was incorporated at three inclusion levels as feed replacement in the basal diet, reflecting different levels of concentrate replacement: 30% (50% hay/35% concentrate/15% feed replacement), 50% (50% hay/25% concentrate/25% feed replacement), and 70% (50% hay/15% concentrate/35% feed replacement). EXP 1 had 8 experimental treatments while EXP 2 had 22 treatments. The fermentation bottle was considered the experimental unit. For each treatment, three independent fermentation bottles were incubated per experimental run, and the experiment was repeated over three separate weeks, resulting in a total of nine replicates per treatment (n = 9).

2.4. In Vitro Incubation, Sample Collection, and Analysis

The procedures for the in vitro batch culture applied in this study followed the protocol described previously by Ahmed et al. (2024) [23]. About 500 mg of the substrate was added to pre-weight nylon bags (BG0510, Sanshin Industrial Co., Ltd., Yokohama, Japan; pore size: 53 ± 10 µm), which were then heat sealed. These bags were placed into 120 mL glass fermentation bottles. The fermentation bottles were flushed continuously with CO₂, while adding 40 mL of artificial saliva prepared according to McDougall (1948) [24], and 20 mL of rumen fluid. Bottles were sealed with rubber and aluminum caps (Maruemu Co., Ltd., Osaka, Japan) and incubated in a rotary shaking incubator (Takasugi Corporation, Tokyo, Japan) at 39 °C for 24 h.

After a 24-h incubation period, the total gas produced was measured using a calibrated glass syringe [25]. Headspace gas from each bottle was collected into vacuum tubes (BD Vacutainer, Becton, Dickinson and Company, Franklin Lakes, NJ, USA). The composition of the gas was then analyzed through gas chromatography (GC-8A, Shimadzu Corp, Kyoto, Japan), following the method previously described by Ahmed et al. (2021) [26]. After that, the fermentation bottles were opened, and the pH was immediately measured with a pH meter (LAQUA F-72, HORIBA Scientific, Kyoto, Japan). From each bottle, 1 mL of culture medium was collected, transferred to an Eppendorf tube (Eppendorf AG, Hamburg, Germany), and immediately centrifuged at 16,000× g for 5 min at 4 °C. The resulting supernatant was diluted threefold with distilled water, filtered through a 0.45 µm cellulose acetate syringe filter (DISMIC—13CP, ADVANTEC, Toyo Roshi Kaisha, Ltd., Tokyo, Japan), and used for further analysis of volatile fatty acids (VFA) using high-performance liquid chromatography (Shimadzu LC-20 HPLC, Shimadzu Corp., Kyoto, Japan), following the method described by Ahmed et al. (2021) [27]. Briefly, VFA were separated using a Shim-pak column (8.0 mm i.d. × 300 mm; Shimadzu Corp., Kyoto, Japan) maintained at 40 °C, with an organic acid mobile phase at a flow rate of 0.8 mL/min, and detected using a conductivity detector (CDD-10AVP, Shimadzu Corp., Kyoto, Japan). Quantification was performed using an external standard method. Afterwards, the nylon bags containing feed residue were washed with tap water until the effluent was clear. They were then dried in an oven at 60 °C for 48 h and re-weighed to determine the IVDMD.

In EXPERIMENT 2, a 0-h wash step was performed before the 24 h incubation to address any potential overestimation of digestibility by quantifying the amount of sample that leaches from the nylon bags before fermentation commences. This measurement is crucial as it prevents inflated values in terms of dry matter disappearance and gas or CH₄ yield (mL/g) of digested (D.DM), particularly when working with fine powders or soluble ingredients. At the start of the 0 h, each sample was weighed, placed into a nylon bag, and sealed. The sealed nylon bags were then submerged in a water bath at 37 °C for 5 min. After removal, the bags were oven-dried at 60 °C for 48 h and re-weighed [28]. The difference between the initial and final weights was used to correct and adjust all digestibility calculations. This procedure was applied only in EXP 2 because the test materials were included in the nylon bags as feed replacements, whereas in EXP 1, the additives were added directly to the fermentation bottles not to the nylon bags.

2.5. Chemical Analysis

The analyses of the chemical composition of the Klein grass hay, concentrate mixture, and the tested samples were conducted following the procedures set forth by the Association of Official Analytical Chemists (AOAC, 1995) [29]. To determine dry matter (DM) content, the samples were placed in aluminum cups and dried in a drying oven at 135 °C for 2 h (method 930.15). The OM and ash content were analyzed by placing samples in crucibles and igniting them in a muffle furnace (DR201, Yamato Scientific, Tokyo, Japan) at 600 °C for 2 h, according to method 942.05. Ether extract (EE) was determined using the method 920.39, using a Soxhlet. Nitrogen (N) content was measured using the Kjeldahl method (method 984.13) with an electric heating digester (DK 20, VELP Scientifica, Usmate (MB), Monza, Italy), and CP was calculated as N multiplied by 6.25. The NDF, ADF, and ADL contents were determined as residual ash values, utilizing an ANKOM Technology Crop system (Macedon, NY, USA). For NDF measurement, sodium sulfite was used without the addition of heat-stable alpha-amylase, employing ANKOM Technology’s fiber analysis system. The non-fiber carbohydrate (NFC) was calculated using this formula: NFC g/kg = 1000 − (NDF g/kg + CP g/kg + EE g/kg + Ash g/kg). The chemical compositions of the experimental treatments in EXP. 2 are shown in

Table 2. Chemical composition (% in dry matter) of the experimental diets containing different levels of the test ingredients as a concentrate mixture replacement in the basal diet (EXP. 2).

Parameter		Dry Matter	Organic Matter	Crude Ash	Crude Protein	Ether Extract	NDF 1	ADF 2	ADL 3	NFC 4
Control		90.98	92.95	7.05	14.41	2.39	44.33	21.00	3.22	31.83
EGY	30%	91.83	93.19	6.81	13.34	2.40	41.34	20.04	3.07	36.12
	50%	92.41	93.36	6.65	12.63	2.41	39.35	19.41	2.98	38.98
	70%	92.98	93.52	6.48	11.92	2.41	37.35	18.78	2.88	41.84
EGG	30%	92.37	92.90	6.80	15.43	3.43	41.09	20.02	3.09	33.25
	50%	93.31	92.87	6.63	16.11	4.13	38.93	19.37	3.00	34.20
	70%	94.24	92.83	6.47	16.79	4.82	36.77	18.72	2.91	35.15
AA	30%	91.72	92.97	7.03	14.72	2.97	41.26	19.96	3.02	34.02
	50%	92.22	92.98	7.02	14.93	3.36	39.22	19.28	2.89	35.48
	70%	92.71	93.00	7.00	15.13	3.75	37.17	18.59	2.75	36.95
AC	30%	92.45	92.43	7.57	13.61	5.41	41.21	19.97	3.02	32.20
	50%	93.44	92.08	7.92	13.08	7.42	39.14	19.28	2.89	32.45
	70%	94.43	91.73	8.27	12.55	9.43	37.06	18.59	2.75	32.70
AD	30%	92.51	92.45	7.55	13.68	5.22	41.23	19.97	3.02	32.32
	50%	93.53	92.11	7.89	13.20	7.11	39.17	19.29	2.89	32.64
	70%	94.55	91.78	8.22	12.71	8.99	37.11	18.61	2.75	32.97
UP	30%	92.21	87.33	12.67	13.19	2.42	42.05	22.24	3.24	29.67
	50%	93.03	83.58	16.42	12.39	2.44	40.52	23.07	3.26	28.24
	70%	93.85	79.84	20.16	11.58	2.46	39.00	23.90	3.27	26.80
GM	30%	92.24	92.42	7.58	13.36	3.13	47.52	24.74	5.82	28.42
	50%	93.08	92.07	7.94	12.66	3.62	49.65	27.24	7.56	26.14
	70%	93.92	91.71	8.29	11.96	4.11	51.77	29.73	9.30	23.87

¹ NDF: neutral detergent fiber. ² ADF: acid detergent fiber. ³ ADL: acid detergent lignin. ⁴ NFC: non-fiber carbohydrate. EGY: Euglena gracilis yellow. EGG: Euglena gracilis green. AA: Aurantiochytrium A. AC: Aurantiochytrium C. AD: Aurantiochytrium D. UP: Undaria pinnatifida. GM: grape marc.

.

2.6. Statistical Analysis

All data were analyzed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). For all experiments, data were analyzed using PROC MIXED models, with treatment included as a fixed effect and experimental run included as a random effect to account for variation among incubation batches. The statistical model used was: Yij = μ + Ti + Rj + εij, where Yij is the dependent variable, μ is the overall mean, Ti is the fixed effect of treatment, Rj is the random effect of experimental run, and εij is the residual error. Least-square means were compared using Tukey’s test. Results are presented as means with pooled standard errors of the mean (SEM). Statistical significance was declared at p < 0.05, and trends were discussed when 0.05 ≤ p < 0.10.

3. Results

3.1. Experiment 1

The addition of materials tested at a 5% inclusion level did not significantly affect the total gas yield (mL/g) of DM or D.DM (

Table 3. Effect of feed additives (5%) on in vitro gas production profile.

Parameter	Control	EGY	EGG	AA	AC	AD	UP	GM	SEM	p-Value
Total gas/DM 1 (mL/g)	46.02	43.56	42.22	43.11	40.44	41.67	40.67	40.44	0.55	0.102
Total gas/D.DM 2 (mL/g)	211.71	193.20	188.34	201.52	191.90	195.85	209.17	201.52	3.20	0.275
CH4 (%)	4.84 a	4.86 a	5.03 a	4.73 a	3.90 ab	4.62 ab	4.27 ab	3.04 b	0.20	0.007
CO2 (%)	95.16 b	95.14 b	94.97 b	95.27 b	96.10 ab	95.38 ab	95.74 ab	96.96 a	0.20	0.007
CH4/DM (mL/g)	2.29 a	2.12 a	2.15 a	2.06 a	1.58 ab	1.91 a	1.73 ab	1.19 b	0.09	0.001
CH4/D.DM (mL/g)	10.42 a	9.36 ab	9.41 a	9.52 a	7.45 ab	9.06 ab	8.72 ab	5.91 b	0.42	0.006
CO2/DM (mL/g)	43.71	41.44	40.07	41.05	38.86	39.75	38.93	39.25	0.51	0.249
CO2/D.DM (mL/g)	201.3	183.84	178.94	192.00	184.45	186.79	200.46	195.61	3.13	0.262

¹ DM: dry matter. ² D.DM: digestible dry matter. EGY: Euglena gracilis yellow. EGG: Euglena gracilis green. AA: Aurantiochytrium A. AC: Aurantiochytrium C. AD: Aurantiochytrium D. UP: Undaria pinnatifida. GM: grape marc. SEM: standard error of the mean. ^a,b Values in the same row with different superscripts differ significantly (p < 0.05).

). The proportions of CH₄ and CO₂ were significantly affected by the addition of GM at 5% as an additive, as indicated by different superscript letters in Table 3. The CH₄ yield (mL/g D.DM) was decreased in GM compared to the control, with a 43.3% reduction. AC also showed a numerical reduction in CH₄ yield (mL/g D.DM) of 27.2%, although this difference was not statistically significant compared to the control. The total CO₂ yield (mL/g) of DM or D.DM was not significantly affected (p = 0.249 and p = 0.262).

Rumen fermentation characteristics were largely unaffected by the additives (

Table 4. Effect of feed additive (5%) on in vitro rumen fermentation characteristics.

Parameter	Control	EGY	EGG	AA	AC	AD	UP	GM	SEM	p-Value
pH	6.58	6.59	6.59	6.60	6.61	6.59	6.58	6.59	0.01	0.544
IVDMD 1 (%)	44.41 a	45.35 a	45.09 ab	43.98 ab	43.56 ab	43.76 ab	41.64 b	42.38 ab	0.38	0.032
Acetate (mM)	38.06	35.94	37.37	36.59	35.9	36.14	35.98	34.34	0.44	0.074
Propionate (mM)	14.07	13.49	13.82	13.88	13.37	13.51	13.61	13.47	0.22	0.549
Butyrate (mM)	5.8	5.31	5.66	5.67	5.46	5.62	5.4	5.56	0.07	0.091
Total VFA 2	57.92	54.74	56.85	56.14	54.72	55.27	54.99	55.36	0.68	0.138
Acetate (%)	65.73	65.65	65.79	65.15	65.63	65.43	65.41	65.71	0.17	0.331
Propionate (%)	24.26	24.58	24.24	24.66	24.33	24.29	24.71	24.22	0.13	0.481
Butyrate (%)	10.01	9.77	9.97	10.18	10.04	10.27	9.88	10.07	0.13	0.051
A/P 3 ratio	2.71	2.67	2.72	2.65	2.71	2.7	2.65	2.73	0.02	0.458

¹ IVDMD: in vitro dry matter digestibility. ² VFA: volatile fatty acid. ³ A/P: acetate/propionate. EGY: Euglena gracilis yellow. EGG: Euglena gracilis green. AA: Aurantiochytrium A. AC: Aurantiochytrium C. AD: Aurantiochytrium D. UP: Undaria pinnatifida. GM: grape marc. SEM: standard error of the mean. ^a,b Values in the same row with different superscript differ significantly (p < 0.05).

). Rumen pH remained stable across treatments, ranging from 6.58 to 6.60, with no significant differences observed among treatments (p = 0.544). The IVDMD showed a significant effect among treatments (p = 0.032), with UP showing the lowest value (p < 0.04) compared with the control. The VFA concentrations and individual VFA concentrations were not significantly changed when compared with the control group (p > 0.05). Additionally, the acetate-to-propionate ratio remained consistent across treatments.

3.2. Experiment 2

As shown in

Table 5. Effect of concentrate mixture replacement with different feed resources on in vitro gas production profile.

Parameter		Total Gas/DM 1	Total Gas/D.DM 2	CH4%	CO2%	CH4/DM	CH4/D.DM	CO2/DM	CO2/D.DM
		(mL/g)	(mL/g)			(mL/g)	(mL/g)	(mL/g)	(mL/g)
Control	0%	44.67	227.10	4.78	95.22	2.13	10.90	42.54	216.20
EGY	30%	42.72	207.04	4.38	95.61	1.87	9.13	40.85	197.91
	50%	37.00 ***	166.26 ***	4.34	95.66	1.61 *	7.14 ***	35.39 ***	159.12 ***
	70%	30.67 ***	134.58 ***	4.04	95.96	1.23 ***	5.39 ***	29.44 ***	129.19 ***
EGG	30%	41.44	184.68 **	4.88	95.12	2.02	8.99	39.43	175.69 **
	50%	39.33 **	162.34 ***	4.68	95.32	1.84	7.64 **	37.50 **	154.71 ***
	70%	34.28 ***	133.32 ***	4.32	95.68	1.49 **	5.80 ***	32.79 ***	127.52 ***
AA	30%	46.47	249.44	4.91	95.09	2.27	12.26	44.19	237.19
	50%	44.88	214.99	5.08	94.91	2.27	10.83	42.61	204.16
	70%	41.94	212.63	4.42	95.58	1.85	9.44	40.09	203.18
AC	30%	40.22	215.63	4.09	95.90	1.64	8.96	38.57	206.67
	50%	36.44 ***	189.98 *	3.96	96.03	1.44 ***	7.71 **	35.00 ***	182.27 *
	70%	32.44 ***	172.46 ***	3.58 *	96.42 *	1.16 ***	6.39 ***	31.28 **	166.08 ***
AD	30%	43.16	234.32	3.95	96.04	1.70	9.57	41.46	224.76
	50%	40.11 *	204.79	3.91	96.08	1.59 *	8.39	38.52	196.39
	70%	35.94 ***	186.52 **	4.19	95.80	1.51 **	7.91 *	34.43 ***	178.61 *
UP	30%	42.44	241.94	3.74	96.26	1.60 *	9.29	40.85	232.65
	50%	39.56 **	212.7	4.05	95.95	1.60 *	8.82	37.96 *	203.88
	70%	36.5 ***	192.87	4.08	95.92	1.49 **	7.97 *	35.02 ***	184.9
GM	30%	42.33	253.24	4.21	95.78	1.78	10.83	40.54	242.41
	50%	37.61 ***	261.08	4.34	95.66	1.64	11.44	35.97 ***	249.63
	70%	35.67 ***	265.78 *	4.58	95.42	1.63	12.27	34.04 ***	253.51 *
SEM		0.339	3.492	0.057	0.057	0.029	0.214	0.320	3.304
p-value		<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

¹ DM: dry matter. ² D.DM: digestible dry matter. EGY: Euglena gracilis yellow. EGG: Euglena gracilis green. AA: Aurantiochytrium A. AC: Aurantiochytrium C. AD: Aurantiochytrium D. UP: Undaria pinnatifida. GM: grape marc. SEM: standard error of the mean. Asterisks mean a significant difference between this level and the control group, * (p < 0.05), ** (p < 0.01), and *** (p < 0.001).

, as a general rule, total gas yield (mL/g) of DM or D.DM declined significantly as the replacement level increased from 30 to 70% (p < 0.001). For the total gas yield (mL/g) of DM, significant reductions were observed at the 50 and 70% replacement levels for all tested feeds except AA, which did not differ from the control. At these levels, EGY, AC, and GM had the greatest reductions (p < 0.001). The EGG and UP showed significant decreases at 50% (p < 0.01) and 70% (p < 0.001) inclusion. The AD had a smaller but significant reduction at 50% (p < 0.05) and a pronounced reduction at 70% (p < 0.001). When expressed per unit of D.DM, total gas yield was significantly reduced by concentrate replacement for all feeds except AA and UP. The EGG showed significant reductions across all inclusion levels (30–70%), while EGY and AC had significant decreases at 50% and 70%. AD showed a significant reduction only at 70%. In contrast, GM showed the opposite response, with total gas yield (mL/g D.DM) increasing with higher inclusion, and reaching significance at 70% (p < 0.05).

The CH₄ proportion in total gas generally declined with increasing inclusion level across treatments, except for GM, and differed significantly from the control only for AC at 70% replacement (p < 0.05). The CH₄ yield (mL/g DM) was significantly reduced for EGY, EGG, AC, AD, and UP, with the greatest reductions at higher replacement levels. Similarly, CH₄ yield (mL/g D.DM) was significantly reduced in the same treatments. The EGY, EGG, and AC showed significant reductions at both 50 and 70% inclusion (p < 0.01 to p < 0.001), whereas AD and UP differed significantly from the control only at 70% inclusion (p < 0.05). Based on CH₄ (mL/g D.DM), EGY and EGG achieved reductions of approximately 30–50% at 50–70% replacement levels, while AC achieved reductions of about 29–41% over the same range; AD and UP showed smaller but significant decreases of approximately 28% and 27%, respectively, at 70% replacement level when compared with the control group. Total CO₂ production decreased with higher replacement levels, but CO₂ (mL/g) of D.DM declined significantly in EGY, EGG, and AC at 50 and 70% inclusion levels, and in AD at 70%, whereas GM again displayed the opposite trend.

The fermentation characteristics presented in

Table 6. Effects of concentrate mixture replacement with different feed resources on rumen fermentation.

Parameter		pH	IVDMD 1 (%)	Acetate	Propionate	Butyrate	Total VFA 2	Acetate (%)	Propionate (%)	Butyrate (%)	A/P 3 Ratio
				(mmol/L)	(mmol/L)	(mmol/L)
Control	0%	6.56	41.56	33.81	13.91	5.50	53.23	63.55	26.12	10.35	2.43
EGY	30%	6.57	44.69	32.98	12.67 ***	5.44	51.09	64.60	24.71 ***	10.69	2.62 ***
	50%	6.43	48.31 **	31.66 ***	11.69 ***	5.15	48.49 ***	65.31 ***	24.02 ***	10.67	2.72 ***
	70%	6.58	49.33 ***	30.40 ***	10.57 ***	4.71 *	45.68 ***	66.56 ***	23.09 ***	10.35	2.89 ***
EGG	30%	6.56	47.27 *	32.97	12.80 ***	5.75	51.52	64.00	24.78 ***	11.22	2.59 **
	50%	6.61	50.56 ***	32.37	11.86 ***	5.47	49.70 ***	65.16 **	23.81 ***	11.03	2.74 ***
	70%	6.62	52.99 ***	31.03 ***	10.83 ***	5.17	47.03 ***	65.98 ***	22.99 ***	11.03	2.87 ***
AA	30%	6.56	40.56	33.85	13.83	5.56	53.24	63.56	25.95	10.48	2.45
	50%	6.57	45.07	34.51	13.71	5.64	53.86	64.01	25.47	10.53	2.52
	70%	6.55	43.74	33.51	13.00 **	5.19	51.71	64.82	25.13 *	10.05	2.59 *
AC	30%	6.58	40.58	32.49	12.46 ***	5.23	50.18 **	64.78	24.76 ***	10.46	2.62 ***
	50%	6.58	42.27	31.96 **	11.72 ***	5.00	48.68 ***	65.66 ***	24.04 ***	10.30	2.73 ***
	70%	6.58	43.47	31.48 ***	10.74 ***	4.62 *	46.85 ***	67.19 ***	22.89 ***	9.92	2.94 ***
AD	30%	6.51	42.50	33.09	12.81 ***	5.50	51.40	64.41	24.81 ***	10.77	2.60 **
	50%	6.55	45.20	32.42	11.94 ***	5.23	49.58 **	65.38 **	24.04 ***	10.59	2.72 ***
	70%	6.33	46.02	31.78 ***	11.00 ***	4.91	47.69 ***	66.65 ***	23.02 ***	10.33	2.89 ***
UP	30%	6.55	39.03	33.16	12.76 ***	5.52	51.44	64.44	24.77 ***	10.79	2.60 **
	50%	6.52	41.40	32.64	12.03 ***	5.23	49.89 ***	65.42 ***	24.07 ***	10.51	2.71 ***
	70%	6.53	42.48	32.35	11.24 ***	4.98	48.57 ***	66.59 ***	23.12 ***	10.28	2.88 ***
GM	30%	6.54	36.74	33.51	12.08 ***	5.46	51.06	65.65 ***	23.62 ***	10.74	2.78 ***
	50%	6.57	32.46 ***	32.92	10.79 ***	5.49	49.20 ***	66.99 ***	21.94 ***	11.06	3.05 ***
	70%	6.58	30.57 ***	33.07	10.12 ***	4.56 **	47.75 ***	69.31 ***	21.13 ***	9.56	3.28 ***
SEM		0.012	0.547	0.205	0.119	0.042	0.338	0.108	0.102	0.067	0.016
p-value		0.271	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001	0.006	<0.001

¹ IVDMD: in vitro dry matter digestibility. ² VFA: volatile fatty acid. ³ A/P: acetate/propionate. EGY: Euglena gracilis yellow. EGG: Euglena gracilis green. AA: Aurantiochytrium A. AC: Aurantiochytrium C. AD: Aurantiochytrium D. UP: Undaria pinnatifida. GM: grape marc. SEM: standard error of the mean. Asterisks mean a significant difference between this level and the control group, * (p < 0.05), ** (p < 0.01), and *** (p < 0.001).

show that ruminal pH was not significantly affected by concentrate replacement level or feed type (p = 0.271), with values ranging from 6.33 to 6.62 across treatments. In contrast, IVDMD was significantly influenced by feed type and replacement level. The IVDMD generally increased with higher inclusion levels (30–70%) for Euglena gracilis treatments. The EGY showed significantly higher IVDMD at 50% (p < 0.01) and 70% (p < 0.001) replacement levels compared with the control, while EGG had consistently higher IVDMD at all inclusion levels, with significance at 30% (p < 0.05) and at 50 and 70% (p < 0.001). The three Aurantiochytrium strains showed only numerical changes in IVDMD, with no statistically significant differences from the control. The UP reduced IVDMD at 30% and 50% replacement, whereas at 70% replacement, IVDMD was numerically higher than control, though the difference was not statistically significant. The GM showed a progressive reduction in IVDMD with increasing inclusion, with significant decreases at 50 and 70% (p < 0.001). Total VFA concentrations decreased as the replacement level increased to 70% (p < 0.001). Significant reductions in total VFA were observed for EGY, EGG, AD, UP, and GM at 50% and 70% inclusion levels, while AC showed significant reductions at all levels. Concentrations of individual VFAs generally declined with higher replacement. As the level increased, the molar proportion of acetate increased and propionate decreased, resulting in higher acetate-to-propionate (A:P) ratios. The highest A:P ratio was at 70% replacement with GM.

4. Discussion

This study found that the chemical composition of the feed resources showed a wide range of nutritional properties, consistent with previous reports and highlighting clear functional differences. Euglena gracilis biomass had a high CP content in EGG (254 g/kg DM). Both EGG and EGY had elevated NFC concentrations (484 and 513 g/kg DM) and low fiber fractions, with NDF values of 125 and 301 g/kg DM. These fiber levels were much lower than those in Klein grass (648 g/kg DM) and were accompanied by very low ADF and lignin contents compared to the basal diet. This supports classifying Euglena gracilis as a highly digestible, non-fibrous microalgal substrate [10,30]. In contrast, AC and AD had high EE contents (234 and 224 g/kg DM), similar to values reported by Gao et al. (2013) [15], indicating significant accumulation of long-chain PUFA, which can affect rumen fermentation and CH₄ formation [31]. The UP had a high ash content (449 g/kg DM) and moderate NDF (343 g/kg DM), consistent with previous findings [32]. Its mineral-rich composition may contribute to the dilute fermentable organic matter typical of brown macroalgae [33]. The GM had high NDF (453 g/kg DM), ADF (335 g/kg DM), and lignin (197 g/kg DM), matching published values and reflecting its condensed tannin content, which limits microbial activity and digestibility [34,35]. These compositional differences might explain the contrasting fermentation responses observed in Experiments 1 and 2.

These compositional contrasts were reflected in distinct rumen fermentation responses, depending on inclusion level and feeding strategy. As a feed additive (EXP 1; 5%), rumen pH, total VFA concentration, VFA proportions, and the acetate-to-propionate ratio remained largely unchanged across treatments, which indicates that energy supply for milk fat (acetate) and lactose (propionate) synthesis would likely be preserved when these materials are used as feed additives in dairy farms [36,37]. This is consistent with previous in vitro studies showing that inclusion of alternative feed ingredients or bioactive substrates is often insufficient to alter rumen fermentation patterns when the basal diet remains the primary source of fermentable OM [11,38]. The reduction in IVDMD observed with UP might be more likely to be related to the presence of bioactive compounds rather than the dilution of fermentable OM, given that UP was applied as a feed additive in EXP 1. Brown macroalgae are known to contain phlorotannins and other phenolic compounds that can inhibit rumen microorganisms, particularly those involved in fiber and starch degradation, by binding to fiber and microbial enzymes, thereby reducing IVDMD without markedly altering VFA profiles or rumen pH [39]. In contrast, concentrate replacement in EXP 2 markedly altered rumen fermentation, with effects becoming more pronounced at higher inclusion levels. Total VFA concentrations declined with increasing the replacement levels, indicating reduced fermentation intensity, a response commonly observed when readily fermentable substrates are replaced by feeds richer in fiber, lipids, or minerals [38,40,41]. Similar reductions in total VFA production have been reported for macroalgae, lipid-rich microalgae, and fibrous grape by-products in vitro [11,40,41]. In parallel, propionate declined more than acetate, increasing the acetate-to-propionate ratio and indicating a shift toward a more acetate-oriented fermentation pattern, consistent with the higher fiber content in the diet and reduced fermentability. Such shifts can influence the balance between milk fat and lactose synthesis in dairy cows [36,37]. Comparable decreases in total VFA production and increases in the acetate-to-propionate ratio have been reported for wine lees included in vitro, which is attributed to limited fermentable substrate availability and the inhibitory effects of polyphenolic compounds [41], and for lipid-rich microalgae such as Euglena, this reflects the inhibitory effects of fatty acids on rumen microorganisms [10]. Digestibility responses were feed- and dose-dependent, with micro- and macroalgal feeds generally maintaining or increasing IVDMD. The inclusion of Euglena increased IVDMD in the present study, supporting the finding of Ahmed et al. (2023) [11], who attributed this to the lower fiber content and higher content of digestible nutrients such as degradable protein. Also, this finding was observed previously [10,42]. Meanwhile, GM markedly reduced IVDMD. This response is consistent with the high NDF, ADF, and lignin contents of GM, which constrain microbial degradation, as well as the presence of polyphenols, particularly condensed tannins, that further limit digestibility and rumen microbial activity [20,34]. Such reductions in digestibility would directly decrease the metabolizable energy available for milk production [43]. It is important to note that although the tested feed materials may contain secondary metabolites that could negatively affect ruminal protein degradation, ammonia nitrogen was not measured in this study. Therefore, future investigations assessing protein degradability alongside VFA profiles would provide a more comprehensive understanding of the impacts of these feeds on rumen fermentation dynamics and microbial ecology.

Consistent with these fermentation responses, total gas yield differed clearly between additive and feed replacement strategies, underscoring the importance of inclusion level and the mode of application. In EXP 1, inclusion of the tested ingredients at 5% as additives did not significantly affect total gas yield (mL/g DM or D.DM), indicating that the basal substrate remained the primary driver of fermentation and that the limited contribution of additives was insufficient to influence cumulative gas production. This agrees with Ahmed et al. (2023) [11], who reported that the low-level inclusion of Euglena gracilis and Asparagopsis taxiformis biomass had no effect on total gas yield in vitro. In contrast, EXP 2 showed a progressive decline in the total gas yield (mL/g) of DM as the concentrate mixture was replaced at 30, 50, and 70%, with the strongest reductions occurring at 50 and 70% inclusion. Such reductions imply lower energy release from fermentation, which would be expected to limit milk yield if applied in vivo at 50 and 70% replacement levels [43]. When expressed as per unit of digestible DM, total gas yield continued to decline for all replacement feeds except GM, for which gas per unit of digestible substrate increased with inclusion level, reflecting the severe reduction in IVDMD caused by its high lignin and tannin contents. These responses indicate a reduction in overall ruminal fermentation extent and largely reflect the compositional characteristics of the replacement feeds, given that KG formed the forage base of the diet. Such reductions in total gas yield are consistent with the replacement of readily fermentable substrates by feeds richer in structural fiber, lipids, minerals, or secondary plant compounds, which constrain microbial activity and fermentation extent; high fiber and lignin contents, such as those in GM, further limit microbial degradation and cumulative gas production [38]. Likewise, several microalgal feeds (EGG, AC, and AD) combined relatively low fiber with high EE concentrations, which dilute fermentable OM and suppress rumen microbial activity, resulting in lower gas output [11,38]. In addition, the high ash and low OM content of UP limited the availability of fermentable substrate, a response commonly reported for marine macroalgae in in vitro rumen systems [40,44].

These reductions in total gas yield were accompanied by substantial mitigation of methanogenesis, particularly under the concentrate replacement strategy. Increasing replacement levels of the experimental feeds resulted in a progressive decline in CH₄ yield (mL/g DM and D.DM) in EXP 2, with the greatest reductions observed at 50 and 70% inclusion. At 70% replacement, CH₄ yield was reduced by approximately 29–45% on a DM basis and 27–51% on a D.DM basis for EGY, EGG, AC, AD, and UP relative to the control, indicating marked suppression of methanogenesis. The magnitude of CH₄ reduction observed at higher replacement levels is consistent with a reduction in overall fermentative activity and H₂ availability in the rumen, as reflected by concurrent declines in total VFA and total gas yield. Long-chain fatty acids and PUFA present in lipid-rich feed ingredients, including several microalgae, are known to suppress methanogenesis by inhibiting H₂-producing bacteria, protozoa-associated methanogens, and by redirecting metabolic H₂ away from CH₄ formation [7]. Accordingly, the pronounced CH₄ response observed with several microalgal feeds further agrees with reports that lipid-rich substrates can suppress methanogenic activity through combined microbial inhibition and altered H₂ metabolism [45]. However, these reductions coincided with lower total VFA and gas production, indicating a trade-off between CH₄ mitigation and fermentative energy supply for milk production [46]. In contrast, the absence of a CH₄ response when the tested materials were applied as feed additives (5%; EXP 1), particularly for Euglena-based feeds, suggests that the amount of lipid or other bioactive components supplied was insufficient to meaningfully alter H₂ production or methanogenic activity when the basal diet remained the dominant source of fermentable substrate, consistent with reports showing that CH₄ responses to lipids and algal supplements are strongly dose-dependent and limited at low inclusion levels [11,45]. This also supports the findings of Ahmed et al. (2023) [11], which highlight that Euglena does not have a direct inhibitory effect on methanogens. The GM showed a unique response pattern, whereby CH₄ yield was reduced by an additive at 5% without suppressing total fermentation, but higher replacement levels altered total gas production and digestibility responses. Accordingly, the effects of GM on CH₄ production differed markedly depending on the inclusion level and feeding strategy: as a feed additive (EXP 1), GM reduced the yield of CH₄ (mL/g D.DM) by 43.3% without affecting IVDMD, pH, or total VFA, suggesting a CH₄-suppressing effect that was not driven by a general depression of fermentation, but rather by the more targeted actions of GM polyphenols (e.g., condensed tannins) on methanogens and/or H₂ transfer [3]. Interestingly, when GM replaced the concentrate at higher levels (EXP 2), CH₄ yield (mL/g DM) declined modestly, while CH₄ yield (mL/g D.DM) increased, indicating that CH₄ mitigation was mainly driven by a strong reduction in digestibility rather than by direct methanogenesis inhibition. This interpretation is supported by in vitro evidence showing that replacing fermentable substrates with GM markedly reduces IVDMD, gas production, and total VFA alongside decreases in CH₄ production and yield, reflecting a pronounced constraint on fermentative activity [34,47]. Therefore, GM would be the best anti-methanogenic feed additive targeting methanogens among the tested additives.

The CH₄ mitigation responses observed in the present study can be interpreted in light of established biochemical and microbial mechanisms associated with lipid- and polyphenol-rich feed resources. Lipid-rich ingredients suppress methanogenesis primarily by reducing H₂ availability for CH₄ formation and by exerting inhibitory effects on methanogenic archaea and protozoa, thereby limiting interspecies H₂ transfer [6,7]. In addition, polyphenolic compounds such as tannins and phlorotannins influence rumen H₂ balance through direct antimicrobial effects and by altering microbial fermentation pathways, although these effects are often accompanied by reductions in fiber degradation, depending on inclusion level [3,5]. Furthermore, DHA-rich microalgae have been shown to modulate rumen microbial populations by suppressing protozoa and methanogens and redirecting metabolic H₂ toward alternative sinks [7]. Although microbiome analysis is important for better understanding the role of these feeds in altering the rumen microbial community, the lack of such analysis represents one of the main shortcomings of this study and warrants consideration in future research.

Taken together, the comparison of the two experimental designs highlights that the CH₄ mitigation potential of the tested feed resources strongly depends on both the inclusion level and mode of dietary inclusion (additive vs. concentrate replacement). When applied as feed additives (EXP 1; 5%), most materials preserved rumen fermentation characteristics, with limited effects on total gas production and digestibility, while specific feeds such as GM exerted targeted CH₄-suppressing effects likely driven by polyphenolic inhibition of methanogenesis. In contrast, when used as concentrate replacements (EXP 2), increasing inclusion levels substantially altered rumen fermentation, resulting in reduced total VFA production, gas yield, and digestibility, alongside marked reductions in CH₄ yield. These findings demonstrate that feed additives may offer CH₄ mitigation without compromising fermentation efficiency, whereas high replacement levels (50 and 70%) impose a trade-off between CH₄ reduction and fermentative energy supply. Among the tested feeds, GM appeared most effective as an additive targeting methanogenesis, but less suitable as a high-level concentrate replacement due to its negative effects on digestibility. This synthesis underscores the importance of tailoring feeding strategies according to the intended function of the feed resource, whether that be as a targeted CH₄ inhibitor or as a bulk dietary ingredient.

5. Conclusions

This study demonstrates that the CH₄-mitigating potential of alternative feed resources in dairy cattle depends more on inclusion strategy than on feed type. As a feed additive (5%), GM significantly reduced CH₄ yield up to 43.3% without compromising rumen fermentation, suggesting a targeted antimethanogenic effect. When used as concentrate replacements (30, 50, and 70%), EGY, EGG, and AC produced clear, dose-dependent reductions in CH₄ yield while maintaining or improving digestibility. However, these reductions were accompanied by lower total gas and VFA production, indicating a shift in fermentation pathways and a potential trade-off with fermentative energy supply.

Overall, lipid-rich microalgae appear more suitable as partial concentrate replacements (up to 30%), whereas GM is more effective as a 5% feed additive. These findings highlight the importance of aligning feed composition with inclusion strategy to achieve CH₄ mitigation without compromising rumen function. Nevertheless, several limitations should be acknowledged. As this study was conducted in in vitro conditions, the results may not fully reflect in vivo rumen dynamics. In addition, ammonia nitrogen was not measured, limiting the interpretation of rumen nitrogen metabolism, and rumen microbial populations were not directly analyzed. Therefore, further in vivo studies are required to confirm these effects under practical feeding conditions and to evaluate their implications for animal performance and long-term sustainability.