Changes in Ruminal Dynamics and Microbial Populations Derived from Supplementation with a Protein Concentrate for Cattle with the Inclusion of Non-Conventional Feeding Sources

Abstract

Feed supplementation strategies are essential for optimizing cattle productivity, and the incorporation of non-conventional feed resources may reduce both production costs and environmental impact. This study evaluated the effects of pelletized protein concentrates (including Acacia farnesiana, A. schaffneri, and Agave duranguensis bagasse) on rumen fermentation parameters, microbial communities, and gas emissions. Fistulated bullocks received the concentrate daily, and ruminal contents were collected and filtered before and after supplementation to assess in vitro gas and methane production, pH, and microbial composition using high-throughput sequencing of 16S rRNA and mcrA amplicons. In addition, in situ degradability was evaluated during and after the supplementation period. Supplementation led to a significant (p < 0.05) reduction in degradability parameters and methane production, along with a marked decrease in the abundance of Methanobrevibacter and an increase in succinate-producing taxa. These effects were attributed to the enhanced levels of non-fiber carbohydrates, hemicellulose, crude protein, and the presence of bioactive secondary metabolites and methanol. Rumen microbiota composition was consistent with previously described core communities, and mcrA-based sequencing proved to be a valuable tool for targeted methanogen detection. Overall, the inclusion of non-conventional ingredients in protein concentrates may improve ruminal fermentation efficiency and contribute to methane mitigation in ruminants, although further in vivo trials on a larger scale are recommended.

1. Introduction

Cattle farming is a vital economic activity in many countries worldwide. Between 2017 and 2019, global beef production reached nearly 70 million tons and is projected to increase by an additional 6.3 million tons by 2029. During the same period, milk production exceeded 839 billion liters, with major contributions from developing countries such as Argentina, India, Mexico, and Pakistan [1]. In these regions, seasonal fluctuations in forage quality often lead to nutritional deficiencies and economic losses, necessitating the use of supplemental feeds and protein concentrates to maintain production efficiency. However, these practices can increase production costs and reduce profitability [2,3].

At the same time, cattle-derived products face increasing scrutiny due to their environmental impact. Livestock systems contribute approximately 14.5% of global anthropogenic greenhouse gas emissions, with cattle alone accounting for two-thirds of this total (mainly due to enteric methane produced during rumen fermentation, which represents nearly 30% of the global methane emissions) [4,5]. In addition to its contribution to climate change, methane production also represents a loss of dietary energy for animals, negatively affecting their growth and productivity [6].

The use of non-conventional feed resources has emerged as a promising strategy to address both environmental and economic challenges. These include industrial by-products and other alternative plant materials that are not traditionally used or commercially exploited in livestock diets [7]. Incorporating such feeds can support sustainable livestock systems by reducing environmental impact and input costs. Examples include plant residues, such as Agave spp. bagasse, a waste product of the mezcal industry, an important beverage in Mexico, along with weeds, shrubs, and aquatic macrophytes [8,9]. Among these, certain Acacia species (particularly A. farnesiana and A. schaffneri, commonly known as huizache in Mexico) have shown potential as high-protein and fiber-rich feed ingredients. These trees are abundant, drought-resistant, and often invasive, making them especially valuable in forage-scarce environments [10,11,12]. Previous in vitro studies have shown that protein concentrates enriched with huizache leaves and Agave duranguensis bagasse can reduce methane production, improve VFAs profiles, and enhance microbial protein synthesis in ruminant diets [13]. Nevertheless, the implementation of such novel feeds requires evaluation of their effects on rumen characteristics of animals under different conditions, including grazing, in which studies have been limited [14].

Ruminants rely on a complex anaerobic microbiota in the rumen to digest structural carbohydrates, as they cannot produce the necessary enzymes [15]. The composition and function of this microbiome play key roles in animal performance, feed efficiency, and methane production. Therefore, understanding how dietary interventions reshape the rumen microbial community is essential for improving productivity and reducing environmental impact [16,17]. While methane plays a physiological role in rumen function, its excessive production is energetically inefficient. Thus, optimizing methanogen populations is a critical target for developing methane mitigation strategies [18].

Most microbial studies rely on sequencing the hypervariable regions of the 16S rRNA gene, which provides taxonomic resolution of bacterial and archaeal communities. These methods are increasingly accessible in developing countries and are instrumental in understanding rumen ecology and improving fermentation efficiency [19]. However, methanogens typically represent only 0.3–3% of the total rumen microbiome [20], and 16S amplicon analysis may fail to capture this diversity. As an alternative, sequencing of the mcrA gene (which encodes the alpha subunit of methyl coenzyme M reductase, an enzyme exclusive to methanogens) has been shown to offer a phylogenetically congruent and functionally informative perspective [21,22,23]. Despite its potential, the application of mcrA sequencing in high-throughput rumen microbiome studies is limited.

The objective of this study was to determine the in vitro production of gas and methane from grazed forage by cattle, its in situ degradability, and microbial populations, before and after supplementation with a pelleted protein concentrate containing unconventional ingredients.

Massive amplicon sequencing of the 16S rRNA and mcrA genes was used to assess the changes in microbial composition associated with supplementation. This study provides novel insights by integrating mcrA-based microbial community profiling with gas-production parameters and degradability. To the best of our knowledge, no previous work has combined these approaches to evaluate the effect of supplementation.

2. Materials and Methods

2.1. Study Area

The study was conducted across three institutions: the Experimental Field of the National Institute for Forestry, Agriculture, and Livestock Research (CIRNOC-INIFAP, Durango, Mexico); the Postgraduate Laboratory of the Faculty of Veterinary Medicine and Animal Science, Universidad Juárez del Estado de Durango (FMVZ-UJED); and the Microbial Biotechnology Laboratory at the Graduate, Research, and Technological Development Unit, Instituto Tecnológico de Durango (UPIDET-TecNM/ITD).

2.2. General Description of the Experiment

In this study, two rumen-fistulated Criollo bullocks of approximately 650 kg live weight were used; these animals were maintained on irrigated pastures before, during, and after the supplementation period. All surgical procedures and the management of rumen-fistulated steers were approved by the Animal Protection Committee of the State of Durango and were conducted in accordance with the Official Mexican Standard NOM-062-ZOO-1999 [24].

Before supplementation, the rumen contents were evacuated manually, and the animals were allowed to graze freely for 60 min. The ingested material during this period was collected through the ruminal fistula and consisted of a mixture of grasses: Lolium perenne, L. multiflorum, Festuca arundinacea, Dactylis glomerata, Bromus inermis, and Trifolium repens. The grazed forage was oven-dried at 55 °C for 48 h and ground through a 1 mm mesh (Bear Cat #1A-S, Westerns Land and Roller Co., Hastings, NE, USA) to serve as the substrate for the in situ degradability and in vitro gas production assays described in the next section.

All experiments were first performed before the supplementation period to allow comparison of the results with those obtained after supplemented feeding. To minimize variability, all animals grazed the same paddock throughout both adaptation and in situ essay periods, and sampling was performed within a narrow time frame to ensure consistency. Throughout the experimental period, the pasture consisted of a homogeneous irrigated grass stand at the vegetative stage, as confirmed by routine pasture monitoring conducted at our station.

In vitro assays, pH measurements, and N-NH₃ determinations were performed in triplicate using ruminal contents. The ruminal content was filtered through four layers of cheesecloth to obtain the fermentation fluid for in vitro trials, and 10 mL aliquots were used to assess pH, N-NH_3, and to perform DNA extraction as described below. The first sample was taken before the first day of concentrate feeding, and the second was taken on the first day after the last concentrate feeding at about 7:00 a.m.

2.3. Pelleted Concentrate Formulation and Supplementation Period

The concentrate included conventional ingredients and two non-conventional feed sources: huizache leaves (Acacia farnesiana and Acacia schaffneri) and Agave duranguensis bagasse. The formulation proportions are listed in

Table 1. Proportion of ingredients in the concentrate.

Ingredient	% DM
Distillers dried grains	15
Agave duranguensis bagasse	6
Ground corn	9
Soybean paste	18
Cottonseed flour	15
Wheat bran	17
Huizache (A. farnesiana leaves)	15
Huizache (A. schaffneri leaves)	5

. The chemical composition of the samples is listed in

Table 2. Chemical composition and degradability parameters of pelletized concentrate and grazed forage.

	Pelletized Concentrate	Grazed Forage
DM	96.33 ± 0.033	92.23 ± 0.033
OM	93.26 ± 0.028	85.00 ± 0.058
CP	25.47 ± 0.581	23.17 ± 0.589
EE	3.5 ± 0.001	5.37 ± 0.145
NDF	31.1 ± 0.173	53.30 ± 1.124
ADF	12.03 ± 0.260	34.33 ± 0.549
HEM	19.07 ± 0.433	18.83 ± 0.433
CEL	9.97 ± 0.0333	26.03 ± 0.617
LIG	1.87 ± 0.203	3.57 ± 0.318
TC	64.3 ± 0.608	57.97 ± 1.198
NSC	33.17 ± 0.751	3.182 ± 1.610
A	6.14 ± 1.112	2.29 ± 0.736
B	55.22 ± 0.282	49.95 ± 1.093
Kd	0.045 ± 0.0064	0.070 ± 0.002

Dry matter (DM), organic matter (OM), crude protein (CP), ethereal extract (EE), Neutral Detergent fiber (NDF), Acid Detergent fiber (ADF), hemicellulose (HEM), cellulose (CEL), lignin (LIG), total carbohydrates (TC), non-structural carbohydrates (NSC), water-soluble fraction (A), potentially degradable water-insoluble fraction (B), degradation rate (Kd, %/h).

.

Table A1. Chemical composition of Acacia leaves included in the concentrate formulation.

%	Acacia farnesiana	Acacia schaffneri	Agave duranguensis bagasse
DM	92.1 ± 0.09	85.1 ± 0.55	81.3 ± 1.78
OM	90.1 ± 0.46	91.7 ± 0.001	86.4 ± 0.74
CP	19.2 ± 0.37	22.4 ± 0.03	4.42 ± 0.056
EE	1.7 ± 0.09	3.9 ± 0.20	2.47 ± 0.09
NDF	55.3 ± 0.81	52.7 ± 1.47	43.82 ± 1.005
ADF	41.9 ± 0.40	32.3 ± 0.40	36.22 ± 0340
HEM	13.5 ± 0.40	20.3 ± 1.88	7.6 ± 1.4
CEL	29.5 ± 0.46	23.2 ± 0.37	30.82 ± 0.49
LIG	11.8 ± 0.26	8.2 ± 0.23	0.80 ± 0.52
TC	71.2 ± 0.27	58.8 ± 0.32	78.49 ± 0.6
NSC	15.8 ± 0.94	6.1 ± 1.79	34.67 ± 0.50
TF	2.3 ± 0.25	2.71 ± 0.10	ND
CT	1.56 ± 0.02	1.74 ± 0.08	ND

Dry matter (DM), Organic matter (OM), Crude protein (CP), Ethereal extract (EE), Neutral Detergent fiber (NDF), Acid Detergent fiber (ADF), Hemicellulose (HEM), Cellulose (CEL), Lignin (LIG), Total carbohydrates (TC), Non-structural carbohydrates (NSC), TF = total phenols expressed in gallic acid equivalents; CT = condensed tannins expressed in catechin equivalents; ND = no detected.

(Appendix A) presents the chemical compositions of the Acacia leaves and Agave duranguensis bagasse included in the concentrate formulation.

For the supply of the pelletized concentrate, the animals received a daily portion of a pelletized protein concentrate at 0.2% of their live weight throughout a 10-day adaptation period and a 5-day in situ assay period. Each morning, the bullocks were housed in individual pens for approximately 30 min (7:00 to 7:30 a.m.) and were fed concentrate during this time. After the total consumption of the supplement, the animals were released and allowed to graze freely. The bullocks had access to water at all times.

2.4. In Vitro Fermentation of Grazed Forage

Fermentation fluids were obtained from ruminal contents by filtering through four layers of cheesecloth before and after the supplementation period at about 7:00 a.m. The first fermentation was started before the first day of concentrate feeding, and the second one on the first day after the last concentrate feeding. Sample handling and in vitro fermentation procedures were conducted under identical conditions to avoid bias.

In vitro gas and methane production were assessed by incubating 0.5 g of the grazed forage with 40 mL of fermentation fluid: buffer solution (1:2) in 120 mL anaerobic flasks at 39 °C for 24 h, according to Theodorou et al. [25]. After incubation time, gas was extracted using a 150 mL syringe, and methane was quantified by injecting the gas into a secondary syringe containing 4 mL of 40% NaOH (w/v), following the method of Fievez et al. [26].

2.5. pH and N-NH₃ Measurement

For this purpose, 10 mL aliquots were obtained by filtration of ruminal contents extracted directly through the ruminal fistula before and after the supplementation period.

The pH was determined using a digital potentiometer (HI83141, Hanna Instruments, Woonsocket, RI, USA). Ammoniacal nitrogen concentrations (N-NH₃) were assessed according to Galyean [27].

2.6. In Situ Degradability of Grazed Forage

For in situ degradability, 10 g of grazed forage was placed in nylon bags with dimensions of 10 cm × 20 cm and a 50 ± 10 μm porosity (R1020 Foraje Bags, ANKOM Technology) and then incubated in the rumen of each bullock in duplicate at predefined time points (0, 3, 6, 9, 12, 24, 48, 72, and 96 h). After retrieval, bags were rinsed and dried, and contents were analyzed: dry matter (DM) and crude protein (CP) were calculated according to the methods described by the AOAC [28]: drying (AOAC 925.45) and the Kjeldahl method using the factor N × 6.25 (AOAC 991.20), respectively; neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined using the filter bag technique with a fiber analyzer (ANKOM Technology, Fairport, NY, USA, EE.UU.). Degradation kinetics were modeled using the Ørskov and McDonald Equation:

D = A + B(1 − exp(−Kd × t))

(1)

where “D” is the amount of feed that disappears over time “t” in hours of incubation, “A” is the water-soluble fraction (%), “B” is the potentially degradable water-insoluble fraction (%) and “Kd” is the degradation rate (%/h). The degradability potential is given by A + B and represents the fraction of material that can be dissolved and degraded in the rumen if the time is not a limiting factor; thus, 100 − (A + B) represents the non-degradable fraction [29]. Table 2 presents the dry matter in situ parameters for the pelletized concentrate and grazed forage before supplementation. In situ assays were performed before the 10-day adaptation period and repeated during the last five days of the supplementation period using the same grazed forage as the substrate.

2.7. Microbial DNA Extraction and Sequencing

Rumen contents collected before and after supplementation of each bullock were filtered through four layers of sterile cheesecloth, and aliquots of 10 mL of the fluid obtained were stored at 20 °C. Samples were centrifuged at 20,000× g for 5 min, and 200 mg of the pellet was used for DNA extraction using the ZymoBIOMICS™ DNA Miniprep Kit (Zymo Research, Cat. D4300, Irvine, CA, USA), according to the manufacturer’s instructions. DNA quality and concentration were assessed spectrophotometrically, and 100 ng of DNA per sample was used for PCR in a 25 μL reaction volume (SureCycler 8800, Agilent Technologies, Santa Clara, CA, USA). Primer sequences (including overhang adapter sequences) and amplification conditions are provided in Appendix A (

Table A2. Primers and PCR conditions.

Amplicons	Primers	Sequence (“Overhand” Adapter Included)	Reference	PCR Conditions	Reaction
Hypervariable region V3/V4 of the ribosomal gen 16S	341F	TCG TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG CCT ACG GGN GGC WGC AG	[17]	3 min at 95 °C, 25 cycles (30 s at 95 °C,30 s at 55 °C and 30 s at 72 °C), 5 min at 72 °C	1.5 mM MgCl2; 0.2 mM dATP, dCTP, dGTP, y dTTP; 0.2 M primers; 1.25 IU of Taq DNA polymerase (Promega Corp., Madison, WI, USA)
	805R	GTC TCG TGG GCT CGG AGA TGT GTA TAA GAG ACA GGA CTA CHV GGG TAT CTA ATC C
α region of the functional gen of Methyl Coenzime M reductase	mlas	TCG TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG GGT GGT GTM GGD TTC ACM CAR TA	[21]	10 min at 95 °C, 30 cycles (30 s at 95 °C,45 s at 60 °C and 45 s at 72 °C), 10 min at 72 °C	4 mM MgCl2; 0.5 mM dATP, dCTP, dGTP, y dTTP; 0.25 M primers; 1.25 IU of Taq DNA polymerase (Promega Corp., Madison, WI, USA)
	mrcA-rev	GTC TCG TGG GCT CGG AGA TGT GTA TAA GAG ACA G CGT TCA TBG CGT AGT TVG GRT AGT

). Amplicons targeted the V3–V4 region of the 16S rRNA and mcrA genes.

2.8. Bioinformatic Analysis

Amplicon libraries were prepared according to the Illumina 16S Metagenomic Sequencing Library Preparation Protocol. Paired-end sequencing (2 × 300 bp) was performed on an Illumina MiSeq platform, generating ~100,000 reads per sample. Sequence assembly was performed using FLASH v1.2.11, and quality control was conducted using FastQC v0.12.1 and MultiQC v1.24.1. Trimming and filtering were performed using Trimmomatic v0.32 and Fastp v0.23.4. Only sequences with a minimum Phred score of Q30 were retained.

On average, 91.3% of the reads passed the quality filters. Approximately 6–12% of the sequences remained unclassified at the genus level, depending on the database used. Taxonomic classification was performed using Kraken 2 [30] on the OmicsBox v2.2.4 platform. Microbial diversity was evaluated using the Shannon diversity index.

2.9. Statistical Analysis

To compare the data before and after protein supplementation, a paired t-test was used to assess whether supplementation generated significant changes in the variables of interest. The level of statistical significance was set at p < 0.05. Correlations between microbial genera and fermentation parameters were assessed using Spearman’s rank correlation (p < 0.05). SAS software (version 9.2; SAS Institute Inc., Cary, NC, USA) was used for this purpose.

3. Results

3.1. In Vitro Gas Production, pH, and Ammoniacal Nitrogen Concentration

Table 3. pH, N-NH₃ concentration, and in vitro gas production *.

	Before Supplementation	After Supplementation	Difference (Mean)	t-Value	p-Value	SEM
pH	6.85	6.13	0.7267	4.27	0.0508	0.1703
N-NH3 (mM)	3.89	3.74	0.1575	1.96	0.1885	0.0802
Total gas (mL/g)	60.96	88.34	−27.38	−3.63	0.0682	7.542
CH4 (mL/g)	9.06	4.96	4.099	7.17	0.0189	0.572
CO2 (mL/g)	51.89	83.38	−31.48	−3.89	0.0602	8.096
CO2/CH4	5.85	16.8	−10.95	−6.95	0.0201	1.5767
%CH4	15.14	5.63	9.5066	4.49	0.0463	2.1192

* p-values that indicate statistical differences are in bold (p < 0.05). SEM: Standard error of the mean. Negative values for the difference and t-value indicate an increase after supplementation, while positive values indicate a decrease.

summarizes the average values of ruminal pH, ammoniacal nitrogen concentration (N–NH₃), and in vitro production of total gas, methane, and carbon dioxide before and after the supplementation period with the pelletized concentrate. A marginally significant reduction (p = 0.0508) in ruminal pH was observed, and the methane volume and percentage after supplementation were significantly reduced (p < 0.05). The CO₂/CH₄ ratio increased significantly (p < 0.05). No statistically significant differences were detected in N–NH₃ concentrations or in other in vitro gas production parameters between pre- and post-supplementation samples. It should be noted that in vitro methane determinations may not accurately predict in vivo emissions [31]; however, these data are of interest when used to complement the other variables evaluated and contribute to the deduction of differences induced in the rumen before and after supplementation.

3.2. In Situ Degradability Parameters

Table 4. In situ degradability parameters by component *.

	Supplementation	A (%)	B (%)	Kd (%/h)	Non-Degradable (%)	Degradability Potential (%)
DM	Before	2.3	50	0.07	47.76	52.24
	After	4.1	38.2	0.03	57.69	42.31
	Difference (mean)	−1.805	11.735	0.0369
	t-value	−2.12	3.63	8.89
	p-value	0.1244	0.036	0.003
	SEM	1.0553	3.2312	0.0042
CP	Before	36.3	52.8	0.08	10.85	89.15
	After	39	29.7	0.06	31.31	68.69
	Difference (mean)	−2.645	23.111	0.0212
	t-value	−2.17	4.65	0.98
	p-value	0.119	0.0188	0.3995
	SEM	1.2216	4.9754	0.0216
NDF	Before	3.5	58.9	0.06	37.61	62.39
	After	2.8	45.4	0.03	51.8	48.2
	Difference (mean)	0.6925	13.494	0.03
	t-value	0.71	3.45	5.57
	p-value	0.5288	0.0409	0.0114
	SEM	0.975	3.9092	0.0416
ADF	Before	1.9	57.5	0.06	40.59	59.41
	After	3.4	47.5	0.02	49.16	50.84
	Difference (mean)	−1.468	10.032	0.0376
	t-value	−1.32	1.63	4.22
	p-value	0.2787	0.2008	0.0243
	SEM	1.1129	6.1403	0.0089
HEM	Before	15.7	54	0.57	30.24	69.76
	After	6.5	49.5	0.02	44.05	55.95
	Difference (mean)	9.2548	4.5553	0.0358
	t-value	6.76	1.72	19.32
	p-value	0.0066	0.1845	0.0003
	SEM	1.37	2.6534	0.0019

* Dry matter (DM), crude protein (CP), neutral detergent fiber (NDF), Acid Detergent fiber (ADF), hemicellulose (HEM), water-soluble fraction (A), potentially degradable water-insoluble fraction (B), degradation rate (Kd, %/h). p-values that indicate statistical differences are in bold (p < 0.05). SEM: Standard error of the mean. Negative values for the difference and t-value indicate an increase after supplementation, while positive values indicate a decrease.

shows the in situ degradability parameters for dry matter (DM), crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), and hemicellulose (HEM) before and after supplementation. The soluble fraction (A) remained statistically unchanged (p > 0.05) for all components, except for HEM, which showed a significant reduction. The potentially degradable insoluble fraction (B) of DM, CP, and NDF decreased significantly (p < 0.05). Similarly, the degradation rate constants (Kd) for DM, NDF, ADF, and HEM were significantly lower after supplementation (p < 0.05), indicating a reduced degradation efficiency for the grazed forage components.

3.3. Microbial Population Analysis

Table 5. Main genera detected by massive sequencing of 16S amplicons *.

Genus	Before Supplementation (%)	After Supplementation (%)	Difference (Mean)	t-Value	p-Value	SEM
Xylanibacter	46.31	35.82	13.02	17.94	0.0004	0.7261
Prevotella	18.24	21.08	−4.08	−2.70	0.0738	1.5118
Aristaella	10.59	5.82	3.49	4.72	0.018	0.7397
Enterococcus	7.13	5.75	3.04	2.63	0.078	1.1548
Butyrivibrio	1.96	4.30	−2.39	−5.75	0.0104	0.4154
Treponema	1.31	3.85	−2.69	−7.17	0.0056	0.3752
Segatella	1.21	5.32	−4.42	−3.33	0.0446	1.3246
Ruminococcus	0.89	1.33	−0.32	−0.51	0.6457	0.6285
Selenomonas	0.36	2.30	−2.03	−3.85	0.0309	0.5255
Anaerostipes	0.33	0.37	−0.13	−0.25	0.8213	0.5073
Olsenella	0.21	0.00	0.28	1.70	0.1884	0.1621
Thermococcus	0.52	0.00	0.60	2.66	0.0766	0.2241
Fibrobacter	0.14	0.34	−0.23	−1.00	0.391	0.2325
Hoylesella	0.33	0.39	−0.14	−0.27	0.8058	0.531
Simiaoa	0.30	0.31	−0.31	−1.00	0.391	0.3125
Paraprevotella	0.33	0.43	−0.43	−1.00	0.391	0.4325

* p-values that indicate statistical differences are in bold (p < 0.05). SEM: Standard error of the mean. Negative values for the difference and t-value indicate an increase after supplementation, while positive values indicate a decrease.

and

Table 6. Main genera detected by massive sequencing of mcrA amplicons *.

Genus	Before Supplementation (%)	After Supplementation (%)	Difference	t-Value	p-Value	SEM
Methanobrevibacter	72.67	51.44	20.41	4.37	0.0221	4.6656
Methanomethylophilus	24.63	43.26	−17.82	−3.40	0.0424	5.2384
Methanosphaera	1.27	3.65	−2.38	−3.91	0.0297	0.6088
Candidatus Methanoplasma	0.86	1.50	−0.73	−1.06	0.3663	0.683
Acinetobacter	0.07	0.02	0.05	1.25	0.2999	0.04
Methanobacterium	0.08	0.04	0.06	1.57	0.2142	0.0366
Paenibacillus	0.03	0.01	0.02	1.63	0.201	0.0122
Methanogenium	0.01	0.02	0.00	0.00	1.000	0.0071
Methanoculleus	0.01	0.02	−0.01	−1.41	0.2522	0.0071
Xylanibacter	0.01	0.00	0.01	1.57	0.2152	0.0048
Haloferax	0.28	0.00	0.28	1.00	0.391	0.28

* p-values that indicate statistical differences are in bold (p < 0.05). SEM: Standard error of the mean. Negative values for the difference and t-value indicate an increase after supplementation, while positive values indicate a decrease.

present the most abundant genera identified using the 16S and mcrA amplicon sequences, respectively. Values indicate mean relative abundance (%) considering the total number of reads classified. Taxonomic classification and statistical analysis of microbial communities revealed no significant differences at higher taxonomic levels between pre-supplementation (BS) and post-supplementation (AS) samples. However, significant changes were observed at the genus level.

In the 16S amplicon dataset, the relative abundances of Butyrivibrio, Treponema, Segatella, and Selenomonas increased significantly (p < 0.05) after supplementation, while that of Xylanibacter decreased (p = 0.0004), and the remaining genera remained statistically unchanged. Conversely, mcrA-targeted analysis revealed significant shifts in the methanogenic communities. Specifically, the abundance of Methanobrevibacter decreased, whereas that of Methanomethylophilus and Methanosphaera increased significantly (p < 0.05), suggesting the selective enrichment of methylotrophic methanogens not detected by 16S amplicon profiling.

The Shannon diversity index at the species level is presented in

Table 7. Shannon Index before and after supplementation *.

Amplicon	Before Supplementation	After Supplementation	Difference	t-Value	p-Value	SEM
16S	4.46	4.47	−0.0120	−0.2000	0.8550	0.0596
mcrA	3.5	3.13	0.3651	1.2000	0.3169	0.3048

* p-values indicate statistical differences (p < 0.05). SEM: Standard error of the mean. Negative values for the difference and t-value indicate an increase after supplementation, while positive values indicate a decrease.

. No significant differences (p > 0.05) in alpha diversity were observed between the BS and AS conditions, indicating that the overall microbial diversity was not substantially altered by the concentrate. However, the diversity indices derived from 16S amplicons were consistently higher than those obtained from mcrA sequences.

Figure A1 and Figure A2 (Appendix B) illustrate the taxonomic distribution of microbial communities at different hierarchical levels (phylum to species) and show the proportions of unclassified sequences. Notably, the percentage of unidentified taxa increased as taxonomic resolution increased, especially in mcrA-based profiles.

3.4. Correlation Analysis

Spearman correlation analysis revealed several significant relationships between the microbial genera and gas production parameters. Xylanibacter, the most abundant genus identified, was positively correlated with Enterococcus (SC = 0.94058, p = 0.0005) and negatively correlated with Prevotella (SC = −0.93862, p = 0.0006), Treponema (SC = −0.80502, p = 0.0159), Segatella (SC = 0.79903, p = 0.0174), and Selenomonas (SC = −0.81129, p = 0.0145). However, no direct correlation was observed between the abundance of Xylanibacter and gas production variables.

Methanobrevibacter, the dominant methanogenic genus, showed a strong positive correlation with Aristaella (SC = 0.90337, p = 0.0021) and with methane production (SC = 0.83916, p = 0.0092). In contrast, it was negatively correlated with Methanomethylophilus (SC = −0.99149, p < 0.0001), which was itself negatively correlated with total gas production (SC = −0.77159, p = 0.0249). Methanosphaera, another methylotrophic methanogen, was positively correlated with Butyrivibrio (SC = 0.94726, p = 0.0004).

Methane production was significantly negatively correlated with the abundance of Butyrivibrio (SC = −0.83488, p = 0.0099), Treponema (SC = −0.84247, p = 0.0087), and Thermococcus (SC = −0.71282, p = 0.0472). It was also inversely correlated with the methylotrophic methanogens Methanomethylophilus (SC = −0.80400, p = 0.0162) and Methanosphaera (SC = −0.75386, p = 0.0492), suggesting that the proliferation of these genera may contribute to lower methane production via competition for hydrogen.

4. Discussion

4.1. Effects on In Vitro Gas Production, pH, and Ammoniacal Nitrogen Concentration

The results shown in Table 3 can be attributed to the chemical composition of the supplemented feed and its influence on the ruminal microbial community, which is further discussed below. Notably, the pelleted concentrate contained higher levels of non-structural carbohydrates (Table 2), which are readily fermentable and have been shown to enhance total gas production in the rumen [3]. Moreover, concentrate-based diets have been associated in previous reports with lower ruminal passage rates and increased propionate production, providing a hydrogen sink that competes with methanogenesis and thereby reduces methane output [6]. In this study, the inclusion of Agave bagasse provided a source of non-forage fiber, which is low in LIG and high in NSC (Table A1), likely due to the cooking process before mezcal fermentation. This has been previously linked to improved feed digestibility by elevating propionate synthesis and reducing hydrogen availability for methane formation [6]. Additionally, the higher crude protein content of the concentrate likely improved nitrogen utilization and microbial protein synthesis, enhancing ruminal efficiency and reducing the energetic losses associated with methane synthesis [32,33]. The presence of tannins in Acacia spp. (Table A1) is known to modulate microbial communities and suppress methanogenesis, which may have contributed to the observed reduction in methane [34,35]. These effects were supported by the strong positive correlation between methane production and the abundance of Methanobrevibacter (r = 0.83916, p = 0.0092), a hydrogenotrophic methanogen, which was significantly reduced following supplementation.

The use of concentrate-based diets can lower ruminal pH and reduce microbial diversity due to the rapid fermentation of starch and sugar [36,37]. However, in this study, despite a marginally significant reduction in pH values after supplementation (p = 0.0508), the values remained within the optimal range for rumen function (pH 6–7), likely due to the relatively low inclusion rate of the concentrate. Moreover, no significant change was observed in the Shannon diversity index (p > 0.05), indicating that microbial diversity was preserved under these experimental conditions (Table 7).

4.2. Effects on In Situ Degradability Parameters

Conversely, the in situ degradability data (Table 4) indicated a decrease in the degradation of grazed forage after supplementation. A reduction was observed in the degradation potential of all evaluated components, including crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), and hemicellulose (HEM). This likely reflects a microbial shift toward more readily fermentable substrates such as non-structural carbohydrates and hemicellulose. These compounds were enriched in the pellets. Similar microbial adaptations to concentrate-rich diets have been previously reported [38], which are consistent with our findings.

Concentrates are known to enhance voluntary intake by increasing palatability and digestibility, which results in higher degradation rates and slower passage through the rumen [38]. This effect may explain why the concentration of ammoniacal nitrogen remained unaffected, despite the reduction in forage degradability. Additionally, the presence of tannins in Acacia leaves, used in the concentrate [13], may have contributed to the reduced degradability. Tannins have been shown to impair the degradation of proteins, hemicellulose, and cellulose via several mechanisms, including substrate binding, enzyme inhibition, and direct antimicrobial effects [39]. Notably, hemicellulolytic enzymes (secreted extracellularly) are particularly vulnerable to inhibition. This could explain the observed decline in hemicellulose-degrading genera, such as Aristaella and Xylanibacter, which is further addressed in subsequent sections.

4.3. Effects on Microbial Populations

4.3.1. Bacterial Community Shifts Based on 16S rRNA Analysis

Numerous studies have reported that a core group of bacterial and archaeal taxa dominates most rumen environments, with dietary variation accounting for most changes in microbial community composition [15,20,40]. The results of the present study are consistent with this concept, as the most abundant microbial groups detected agreed with those frequently reported in previous rumen analyses, particularly at the higher taxonomic levels.

In the 16S rRNA amplicon data, Bacteroidota and Bacillota (formerly Firmicutes) were the dominant phyla, accounting for approximately 72.6% and 23.4% of the sequences before supplementation and 71.4% and 24.4% after supplementation, respectively. In contrast, mcrA amplicon analysis revealed that Euryarchaeota (75.6% before supplementation, 59% after) and Candidatus Thermoplasmatota (23.4% BS, 24.4% AS) were the predominant archaeal phyla. However, as recognized in recent literature, a substantial proportion of the ruminal microbiome remains poorly characterized; furthermore, several taxa identified in this study (such as the genera Xylanibacter and Aristaella) belong to recently proposed classifications [41,42]. Consequently, these genera are often absent from earlier reports simply due to their more recent taxonomic recognition.

The inclusion of grain-rich concentrates in ruminant diets can lead to changes in microbial populations, typically characterized by an increased abundance of Proteobacteria and a reduced abundance of Firmicutes and Bacteroidetes, reflecting microbial adaptation to more readily fermentable carbohydrate substrates [43]. For instance, members of the genus Bacillus, known for their high proteolytic activity, may increase in abundance under such dietary conditions [44]. Nevertheless, in our study, no significant differences were observed in the abundance of these phyla or genera before and after supplementation, likely due to the low inclusion level (0.2% of live weight/day) of the pelletized concentrate. Despite this, significant changes were detected in specific microbial populations, some of which have been associated with reductions in methane production, as is reported in the present work (Table 3).

As expected, bacterial diversity (Table 7) was higher than archaeal diversity, which is consistent with prior observations [45]. In support of this, Wang et al. [6] emphasized that methanogenic archaeal communities are much less diverse than their bacterial counterparts in the rumen.

Xylanibacter was confirmed as a distinct genus from Prevotella based on phylogenomic analysis; however, it retains conserved metabolic functions with other genera in the Prevotellaceae family, which was the most abundant group detected via 16S amplicon sequencing in this study, consistent with previous reports. Prevotellaceae are known for their involvement in polysaccharide degradation [38]. Specifically, Xylanibacter ruminicola (the predominant species detected before and after supplementation (64.51% BS, 53.93% AS)) can ferment a wide range of carbohydrates, including xylose, arabinose, cellobiose, rhamnose, salicin, sucrose, and lactose [41,46]. Its abundance decreased after supplementation (p = 0.0004).

In contrast, Prevotella was the second most abundant genus, comprising ~5% of the species-level 16S sequences (including Prevotella spp. E15-22, P. corporis, and Prevotella spp. E9-3). This genus encompasses highly diverse bacteria capable of utilizing starch, non-cellulosic polysaccharides, and simple sugars, with succinate being the major fermentation product. Many Prevotella species also exhibit hemicellulolytic and proteolytic activities [41]. The observed negative correlation between Xylanibacter and Prevotella suggests competitive interactions, potentially influenced by the higher availability of NSC and hemicellulose in the supplemented diet. This may have favored Prevotella, which exhibits a broader metabolic plasticity [47].

Aristaella hokkaidonensis, the type species of the Aristaella genus, ranked fourth in abundance (2.43% BS, 1.30% AS; p = 0.018). As described by Mahoney-Kurpe et al. [42], Aristaella metabolizes hemicellulose and pectin into short-chain fatty acids (SCFAs) or volatile fatty acids (VFAs), lactate, hydrogen, and ethanol. A strong positive correlation with Methanobrevibacter abundance suggests that its reduction after supplementation may have decreased hydrogen availability for hydrogenotrophic methanogenesis.

Among Firmicutes (or Bacillota), the second most abundant phylum, Lachnospiraceae was the dominant family and significantly increased after supplementation (4.54% BS to 6.89% AS; p < 0.05). This family includes the genera Butyrivibrio and Pseudobutyrivibrio, both of which are involved in butyrate synthesis [44]. Notably, Pseudobutyrivibrio xylanivorans was detected only after supplementation (1.2%), and the abundance of Butyrivibrio increased significantly (p =0.0104) (Table 5). Members of this family are key fibrolytic microbes capable of degrading xylan, xyloglucan, and pectin and producing butyrate, acetate, and H₂. Some also generate methanol, which serves as a substrate for methylotrophic methanogens such as Methanosphaera [15,40]. The observed positive correlation between Lachnospiraceae and Methanosphaera (SC 0.94726) supports this hypothesis. However, overall methane production declined post-supplementation, suggesting that hydrogenotrophic methanogens (e.g., Methanobrevibacter) may have a more substantial role in methanogenesis than methylotrophs, as previously reported [48], and aligns with findings by Henderson et al. [15] and Xie et al. [49], who noted that Butyrivibrio thrives in diets rich in NDF and crude protein provided by the protein concentrate.

Enterococcus, a lactic acid-producing genus capable of fermenting diverse substrates, was the fourth most abundant genus (7.13% BS, 5.75% AS). Its abundance was positively correlated with Xylanibacter (SC 0.94058, p = 0.0005) and negatively correlated with Prevotella (SC −0.94574), indicating potential synergism and competition, respectively. Enterococcus faecalis (11.24% BS and 8.46% AS) was the dominant species. This genus utilizes H₂ for succinate formation, competes with methanogens, and has probiotic potential [44,45].

Treponema, another fibrolytic genus, also increased significantly (1.31 BS, 3.85% AS; p = 0.0056), likely adapting to the altered polysaccharide profile of the concentrate-fed animals. While Treponema lacks cellulolytic activity, it excels in hemicellulose degradation, facilitating access to fiber for other microbes [49,50,51]. Its abundance was negatively correlated with Xylanibacter (SC −0.80502) and methane (SC −0.84247) and positively correlated with Butyrivibrio, Selenomonas, and total gas production, suggesting metabolic pathways that divert H₂ from methane production.

An increase was also observed in the Negativicutes class (0.81%BS, 2.14% AS; p < 0.05), particularly Selenomonadales (0.12% BS, 0.87% AS), which converts succinate to propionate, thereby reducing H₂ availability for methanogens [52,53,54,55]. This was supported by positive correlations with Treponema (SC 0.94235), Prevotella (SC 0.6866), and Selenomonas, which were significantly increased (p = 0.0309). This genus includes Selenomonas ruminantium (0.36% BS to 3.45%), a key species in lactate fermentation and VFA production [56,57].

Fibrobacter abundance did not significantly change, but it tended to increase, possibly reflecting positive interactions with Treponema [58,59]. Fibrobacter succinogenes, the major fibrolytic species detected (0.26% BS; 0.94% AS; p > 0.05), is known to produce cellulolytic, xylanolytic, and pectinolytic enzymes and contribute to propionate formation [44].

Although the novelty of this study lies in validating the effects of concentrate consumption under practical grazing conditions using both in situ and in vitro approaches, it is important to highlight that, as VFAs were not quantified in the present study, further approaches must be performed to complement the present data, including the measurement of these important compounds.

4.3.2. Methanogenic Archaea Shifts Based on mcrA Analysis

The study of methanogenic archaea has shown that the functional gene mcrA provides a phylogenetic resolution comparable to that of the 16S rRNA gene, as first reported by Luton et al. [23] and supported by other studies [21,22]. In our dataset, more than 99.6% of the mcrA amplicon sequences were assigned to the Archaea domain. Minor fractions of bacteria and eukaryotes, primarily from the Pseudomonadota (Proteobacteria), Firmicutes, and Bacteroidota phyla, were detected but represented less than 0.25% of the sequences.

The archaeal community was dominated by Euryarchaeota and Candidatus Thermoplasmatota phyla, consistent with previous rumen studies across species, locations, and diets [6,15]. The methanogenic community displayed limited diversity, with only a few genera detected: Methanobrevibacter, Methanomassiliicoccus, Methanomicrobium, Methanosphaera, and Methanobacterium. This conserved composition aligns with the observations of Wang et al. [60].

Among hydrogenotrophic methanogens, Methanobrevibacter was dominant but showed a significant decline following supplementation (72.67%BS vs. 51.44% AS, p = 0.0221). Several species were identified, including Methanobrevibacter millerae, which was the most abundant. Although this species has not been widely reported as dominant, previous studies on Tibetan and Indian sheep populations have found it to be prominent [48,61]. This suggests that local environmental and dietary conditions strongly influence the composition of methanogens [15,40,48].

The genus Methanosphaera (which requires methanol as a substrate) significantly increased after supplementation (1.27% BS vs. 3.65% AS, p = 0.0297). Methanol is a by-product of pectin degradation by Treponema and Butyrivibrio, whose relative abundances also increased under the supplemented condition. Additionally, Agave bagasse, included in the supplement, may provide methanol, which can be produced during mezcal production [62]. Although a rise in Methanosphaera abundance was observed, it did not correspond to increased methane production, indicating that methylotrophic methanogens did not offset the overall reduction in methanogenesis. These associations have been previously suggested [15,63].

The second most abundant archaeal class was Thermoplasmata (18.22% BS vs. 43.78% AS, p < 0.05), including Methanomethylophilus and Methanoplasma. Notably, Methanomethylophilus increased significantly (24.63% BS vs. 43.26% AS, p = 0.0424) and was negatively correlated with methane levels (SC = −0.80403, p = 0.0162) and Aristaella (SC = −0.91317, p = 0.0015). This may reflect the competitive or compensatory ecological dynamics of Methanobrevibacter. Members of Methanomassiliicoccales use H₂ as an electron donor to reduce methyl compounds to methane; however, their proliferation did not lead to higher methane emissions, indicating lower methanogenic efficiency.

Our findings are consistent with those of Wang et al. [60], who reported reduced methane production and a shift in archaeal communities after dietary supplementation. Similar effects were observed by Ogata et al. [36], who noted a reduction in Methanobrevibacter abundance. McCabe et al. [64] suggested that this genus competes with succinate-producing bacteria, which were elevated in our study and are known to divert H₂ from methanogenesis. Therefore, shifts in microbial community structure, rather than total methanogen abundance, appear to be critical for methane outcomes [6,64,65].

Furthermore, Methanobrevibacter encodes a sugar-binding protein that allows adherence to rumen protozoa, facilitating the interspecies hydrogen transfer. Protozoa are known targets of condensed tannins, which can disrupt their membranes and interfere with their symbiosis with methanogens [66,67,68]. This may help explain the observed reduction in Methanobrevibacter and total methane production.

Malik et al. [48] reported that hydrogenotrophic methanogens like Methanobrevibacter contribute more significantly (30%) to methane yield than methylotrophic species (19%). This supports the notion that reductions in hydrogenotrophic methanogens, as observed in this study, are more influential in mitigating methane emissions.

Lastly, unidentified populations constituted a large proportion of mcrA and 16S amplicon sequences (20–80% and 6–84%, respectively), underscoring the still-unknown diversity of the rumen microbiome [15,38,59,69]. Continued exploration of this microbial complexity is essential for advancing strategies to enhance ruminant productivity while mitigating its environmental impact.

4.4. Limitations and Future Directions

This study had several limitations. The most critical constraint was the small sample size (n = 2), which limited the statistical power and generalizability of the findings. Therefore, the results should be viewed as preliminary rather than definitive. Moreover, the forage intake composition was not directly measured, and the assumption of uniform intake relied on previous reports; however, variations in forage composition cannot be completely ruled out and may have had some impact on the results. Although sequencing and gas production analyses were carefully conducted, the small number of biological replicates in some molecular assays reduces confidence in microbial abundance estimates. Additionally, volatile fatty acid (VFA) profiles were not measured, precluding direct correlations between microbial shifts and fermentation end-products such as acetate or propionate; thus, the discussion related to VFAs is based on previous research. The use of an in vitro system to assess methane production, while informative, may not fully represent in vivo rumen dynamics under grazing conditions. Even though the experiments were conducted over a short period to minimize bias, seasonal variability in forage quality, a factor known to influence rumen fermentation, was not assessed. Although the application of 16S rRNA and mcrA sequencing provides relevant taxonomic insights, these approaches offer limited functional resolution. Future studies should address these gaps by incorporating larger animal cohorts with biological replication, monitoring VFA profiles, and applying integrative functional approaches, such as shotgun metagenomics or transcriptomics, to better understand the metabolic consequences of microbial shifts in response to dietary interventions.

5. Conclusions

The inclusion of a pelletized protein concentrate formulated with non-conventional feed sources, such as Acacia and Agave bagasse, appeared to influence rumen fermentation patterns and microbial populations in grazing cattle under the specific conditions of this exploratory study. Notably, the reduction in Methanobrevibacter spp., the dominant hydrogenotrophic methanogens, along with the increased abundance of Butyrivibrio, Treponema, and Selenomonas, suggests a microbial shift towards alternative hydrogen sinks, as indicated by the reduction in methane after supplementation. These changes may be associated with the higher levels of non-structural carbohydrates, crude protein, phytochemicals, and methanol present in the concentrate.

In vitro methane production was significantly reduced, indicating promising potential for enteric methane mitigation (estimated at 15–20% per animal/day under the conditions tested). mcrA-based amplicon analysis proved to be a useful molecular tool for monitoring archaeal population dynamics and inferring functional methane-related shifts, avoiding interference from dominant bacterial taxa.

However, these findings should be interpreted as preliminary, and in vivo validation is required to confirm these microbial and functional changes under practical feeding conditions. Future research should assess higher inclusion levels, longer supplementation periods, and larger sample sizes, and be complemented with an analysis of volatile fatty acids and forage intake to address the long-term implications of methane emissions, animal performance, and ruminal health in production systems.