In Vitro Evaluation of Cattle Diets with the Inclusion of a Pelletized Concentrate Containing Acacia farnesiana

Simple Summary

This study evaluated huizache (Acacia farnesiana) leaves as a sustainable option in ruminant diets. Pelleted concentrates with a fixed 10% huizache inclusion were offered at increasing proportions, and their impact on methane production was assessed. The in vitro results showed that huizache did not raise methane output, indicating its potential as a non-conventional forage resource. Further studies under practical feeding conditions are needed to confirm its role in reducing emissions and supporting sustainable livestock systems.

Abstract

Livestock production raises significant environmental concerns, necessitating the development of sustainable feeding strategies based on non-conventional forages, such as locally available vegetation. This study evaluated the effects of a pelleted concentrate containing 10% Acacia farnesiana leaves as a dietary supplement on in vitro ruminal fermentation. Four experimental diets were formulated with increasing levels of the concentrate (0%, 25%, 50%, and 75%). Analyses were performed in triplicate and included chemical composition, in vitro gas and methane production, fermentation kinetics, ammonia nitrogen concentration (N–NH₃), in vitro dry matter digestibility (IVDMD), and metabolizable energy (ME) estimation. The results revealed no significant differences (p > 0.05) in most gas production kinetic parameters, overall fermentation patterns, or metabolizable energy. In contrast, a significant increase (p < 0.05) in secondary metabolite concentrations was detected. While methane production remained unaltered (p > 0.05), a significant linear reduction was observed for IVDMD, the lag phase (L), and N–NH₃ concentration (p = 0.0064, p = 0.0036, and p < 0.0001, respectively). These findings suggest that A. farnesiana can be incorporated into ruminant concentrates without increasing methane emissions. However, in vivo trials and mechanistic studies are required to validate and further elucidate these results.

1. Introduction

Livestock plays an important role in food systems. It is a source of high-quality protein and other nutrients such as vitamins and minerals [1]. According to the Food and Agriculture Organization of the United Nations [2], livestock accounted for 40% of world production and contributed to the subsistence and food security of almost one billion people, providing 15% of the world’s food energy and 25% of the protein in the world’s diets. In addition, 60% of rural households have cattle [3]; however, due to the evident scarcity of conventional forages used for cattle feed, the inclusion of non-conventional forage species in cattle diets has been implemented [4]. The ecosystem of the State of Durango (Mexico) offers alternative sources of potential forages that can be exploited, such as A. farnesiana, which can be used as a low-cost source of nutrients for ruminants [5]. This shrub legume contains tannins, which are anti-nutritional agents [6]; however, tannins can influence ruminal fermentation through protein binding, hydrogen sequestration, and modulation of microbial populations. These effects are often accompanied by trade-offs such as reduced feed intake or fiber digestibility [7]. In addition to its practical advantages for transport and storage, pelleting may also influence the nutritive value of feed ingredients. The densification process can alter nutrient availability, particularly protein and starch fractions, through heat and pressure effects during pellet formation [8]. Considering these potential changes, the evaluation of A. farnesiana in pelleted form provides a more realistic assessment of its use under practical feeding conditions. The objective of this research was to study the effect of the inclusion of a pelleted concentrate with A. farnesiana in cattle diets on its chemical composition, gas production, and in vitro digestibility.

2. Materials and Methods

2.1. Location of Study Area

The experiment was carried out in the Industrial Biotechnology and Technological Innovation Laboratory of the TecNM—Technological Institute of Durango in the Postgraduate Unit, Research and Technological Development (UPIDET) in collaboration with the Faculty of Veterinary Medicine and Animal Husbandry of the Universidad Juárez del Estado de Durango (UJED), in the laboratory of the Postgraduate Unit of the Institution. All assays were performed in triplicate.

2.2. Vegetative Material

Branches of A. farnesiana were collected in Durango, Mexico (23°58′49.7″ N, 104°40′59.0″ W) during the late summer season (August 2023), when plants were in the vegetative–early flowering stage. The plants were not under biotic or abiotic stress conditions. Sampling was conducted randomly by cutting branches approximately 1 m in length; only mature green leaves were selected, while stems and senescent material were excluded. Leaves were manually separated from the branches and carefully shaken and brushed to minimize soil or dust contamination before further processing. The remaining ingredients used in the pelleted concentrates were sourced from local forage suppliers.

2.3. Preparation of the Pelletized Concentrate with A. farnesiana

For the pellets, a formulation with a 14.7% crude protein content was selected [9].

Table 1. Ingredient proportions and their respective contributions to the crude protein content of the pellets.

Ingredient	g kg−1 DM *	Crude Protein (g kg−1 DM *)
Ground corn	600	5.88
Wheat bran	115	1.84
Acacia farnesiana leaves	100	1.71
Sugar cane molasses	60	0.36
Soybean meal	50	2.4
Dried corn distiller’s grains	40	1.2
Cottonseed meal	35	1.4

* Dry matter.

presents the formulation of the pellets, listing the ingredients used. All the ingredients, including A. farnesiana leaves, were dried in a forced-air oven at 55 °C (Calisa Alley Model 550R) until a constant weight was achieved. They were then processed in a mill (Keiter IC60B Mill) and sieved using a 1 mm mesh.

Pelleting

The pelletizing process was carried out in a laboratory-scale pelletizing mill “Romel Commerce” (Mod. PEL107) with a cylindrical inlet diameter of 6 mm, an inlet at room temperature, and a humidity of 35% where the mixture with the ingredients for the pelletized concentrate was introduced. The pellets were left to dry at room temperature for 24 h. They were then dried in a forced-air oven at 60 °C for 24 h and subsequently stored at room temperature until use. Finally, they were added in the aforementioned proportion to each of the treatments.

2.4. Experimental Treatments

Four balanced diets were formulated and adjusted to a 12.3% crude protein content, incorporating the pelleted concentrate with A. farnesiana at four percentages: 0%, 25%, 50% and 75%. The inclusion of A. farnesiana leaves was, in all cases, 10% of the concentrate (Table 1). The nutritional balance was performed using the Oklahoma State University beef ration calculator software [10]. The ingredients and their composition are presented in

Table 2. Ingredient proportions in experimental diets, on dry matter basis.

Ingredient	T0	T1	T2	T3
Pelleted concentrate *	0	25	50	75
Ground corn	64	44	24	0
Alfalfa hay	12.5	10	7	4
Oat hay	12.5	15	18	20
Dried corn distiller’s grains	10	5	0	0
Limestone	1	1	1	1

* With 10% of A. farnesiana leaves.

.

2.5. Chemical Composition of Pelleting Ingredients and Treatments

Dry matter (DM, 930.15), ethereal extract (EE, 920.39), ash (942.05), and crude protein (CP, 990.03) contents were determined in triplicate for each treatment according to standardized procedures by AOAC [11]. Neutral detergent fiber (NDF), acid detergent fiber (ADF), and lignin (LIG) were evaluated as described by Van Soest et al. (1991) [12]. It is important to note that α-amylase was not included during the NDF analysis of ground corn. The absence of α-amylase treatment may have led to partial interference of starch in the NDF fraction, resulting in overestimation of NDF and underestimation of NFC values for this ingredient. Hemicellulose (HEM), and cellulose (CEL) fractions were evaluated with an ANKOM fiber analyzer (Fiber Analyzer 200, ANKOM Technology, Macedon, NY, USA) following the manufacturer’s instructions. Nonstructural carbohydrates (NSC) and total carbohydrates (TC) were calculated according to the following equations [12]:

$$NSC=\left[100-\left(CP+EE+ASH+NDF\right)\right],$$

(1)

$$TC=(NSC+NDF)$$

(2)

2.5.1. Determination of Phenolic Compounds

Extractions were performed by adding 1 g of each sample to 90 mL of an ethanol solution (70% v/v) in plastic vials that were kept under agitation for 24 h. They were filtered and evaporated under vacuum at 40 °C. The extracts were dried and weighed. Condensed tannins (CT) were determined according to the procedures proposed by Heimler et al. [13] using catechin as the standard, and the absorbance was measured in a UV-VIS spectrophotometer at 500 nm (Spectronic Instruments, Genesys 10S, Mod. 336003, Thermo Scientific, Madison, WI, USA). On the other hand, total phenolic compounds (TPC) were determined using the Folin-Ciocalteau method previously described by Singleton and Rossi [14]; gallic acid was used as standard, and absorbance was measured at 760 nm (Spectronic Instruments, Genesys 10S, Mod. 336003).

2.5.2. In Vitro Digestibility of Dry Matter

Dry matter in vitro digestibility (IVDMD) was determined by measuring the dry matter disappearance after 48 h of fermentation at 39 °C using ruminal liquid and buffer solution in a 1:4 ratio. This essay was performed using a Daisy incubator following the manufacturer’s instructions (ANKOM Technology, Macedon, NY, USA). Briefly, 0.5 g of each grounded sample was added to a previously dried to constant weight filter bag (bag: sample ratio (20:1) #F57, 25 micron porosity, ANKOM Technology) and collocated into a Daisy digestion jar containing 1600 mL of the buffer, that consisted of a solution A (KH₂PO₄, 10 g/L; MgSO₄·7H₂O, 0.5 g/L; NaCl, 0.5 g/L; CaCl₂·2H₂O, 0.1 g/L; Urea 0.5 g/L) adjusted to a pH of 6.8 by the addition of the solution B (Na₂CO₃, 15.0 g/L; Na₂S·9H₂O) at 39 °C, and 400 mL of rumen inoculum obtained by filtering the mixed digesta from two stabled fistulated bovines fed with oat hay, alfalfa hay, and ground corn (70:20:10) through four layers of cheesecloth. The container was purged with CO₂ to ensure anaerobic conditions, and digestion took place under agitation and constant temperature for 48 h. After that, filter bags were rinsed with cold tap water, dried, and weighted to determine the dry matter digestibility. Blank corrections were applied, and runs were evaluated in triplicate.

Metabolizable energy (ME) was estimated according to the equation proposed by Menke and Steingass [15]:

$$ME\left({\displaystyle \frac{\mathrm{M}\mathrm{J}}{\mathrm{K}\mathrm{g}\mathrm{M}\mathrm{S}}}\right)=\left[2.20+0.136\left(PG24\mathrm{h}\right)+0.057\left(PC\right)+0.0029{\left(EE\right)}^2\right]$$

(3)

2.6. In Vitro Gas Production

0.5 g of each ground sample was placed into a calibrated 100 mL glass syringe (n = 3 per run, FORTUNA, Düsseldorf, Germany), and 30 mL of an inoculum solution was added. The solution was prepared with rumen fluid obtained by filtering the digesta from two fistulated bovines (fed with 70% oat hay, 20% alfalfa hay, and 10% ground corn) through four layers of cheesecloth and artificial saliva (Na₂HPO₄, 1.4 g/L; KH₂PO₄, 1.5 g/L; MgSO₄·7H₂O, 0.14 g/L; CaCl₂·2H₂O,1.58 mg/L; MnCl₂·4H₂O, 1.2 mg/L; CoCl₂·6H₂O, 0.12 mg/L; FeCl₂·6H₂O, 0.1 mg/L; NaHCO₃, 8.4 g/L; (NH₄)HCO₃, 0.1 g/L; resazurin, 1.22 mg/L; NaOH, 0.08 g/L; Na₂S·7H₂O, 0.336 g/L) at a ratio of 1:2, as described by Theodorou et al. [16]. Measurement of in vitro gas production was quantified by displacement of the syringe plunger at 0, 3, 6, 9, 12, 24, 48, 72, and 96 h to estimate the cumulative total gas production. The gas volume data were fitted to the Gompertz function to obtain the parameters of the kinetics of in vitro gas production according to Equation (4). The Gompertz equation was chosen because it reliably models sigmoidal biological processes, capturing lag, growth, and plateau phases with parameters that have clear biological interpretation [17].

$$PG=A{e}^{-L{e}^{-\left(k{d}^t\right)}}$$

(4)

where PG = volume of gas produced at time t (mL/g DM), A = maximum gas production from the fermentable fraction (mL), kd = constant rate of gas production (mL/h⁻¹), L = latency time before gas production (h).

2.7. In Vitro Methane Production, Carbon Dioxide, and Ammonia Nitrogen (N-NH₃) Concentration

For the quantification of methane and carbon dioxide production, a gas production measurement system (ANKOM RF Gas Production System USA) was used. Glass modules (ANKOM, USA) were tempered to a temperature of 39 °C, then 1 g of sample and 120 mL of a solution prepared with buffer solution and rumen liquid in a 2:1 ratio were added, the modules were incubated in triplicate per sample at 39 °C and left to ferment. Gas production was measured with the ANKOM RF system, where headspace volume was calculated from bottle capacity minus liquid added (e.g., 160 mL in 250 mL bottles). Pressure was recorded with ±1% accuracy (±0.04 psi resolution) and validated using RF70 positive control capsules; gas composition was not measured directly. A single 24-h timepoint was chosen as it represents the standard incubation period commonly used to assess in vitro fermentation end-products, providing a reliable estimate of cumulative gas production while ensuring comparability with previous studies [15]. After 24 h, the pressure valve was opened for 2 s, and the released gas was conducted into a gas analyzer (GEM5000, LANDTEC, Florence, SC, USA) [18]; methane and carbon dioxide readings were taken for future interpretation.

For the quantification of ammoniacal nitrogen, the methodology proposed by Galyean [19] was followed, where a sample of 10 mL was taken from each of the treatments at the end of the fermentation previously described, deposited in plastic flasks provided with 0.3 mL of 50% sulfuric acid as preservative, and kept frozen until analysis. Briefly, 0.5 mL of the sample were mixed with 2.5 mL of a phenol reagent (sodium nitroferricyanide and 90% wt/vol phenol), and 2 mL of a hypochlorite reagent (sodium hydroxide, disodium phosphate and 5.25% sodium hypochlorite) and then heated at 95 °C in a water bath for 5 min and allowed to cool, a reading was made in a spectrophotometer at 630 nm. To calculate the concentration of ammoniacal nitrogen in each sample, a calibration curve was constructed using an ammonia solution of known concentration.

2.8. Statistical Analysis

For chemical composition, in vitro digestibility, and in vitro gas production, a completely randomized design was applied. To detect differences between least square means, Tukey’s multiple range test was used, declaring significant differences at p < 0.05 (SAS 9.1, 2003). For the analysis of in vitro gas production kinetics, an adjustment to the Gompertz model was made using SAS (2003). Considering that the treatment factor had four equidistant categorical levels (0%, 25%, 50%, and 75% concentrate inclusion in the diets), orthogonal polynomial contrasts were used to evaluate significant trends (linear, quadratic, and cubic) in the data for the kinetics gas production parameters and fermentation patterns (SAS 9.1, 2003).

3. Results

3.1. Chemical Composition of the Ingredients Used in the Preparation of Pellets

The results of the chemical composition of the A. farnesiana leaves are presented in

Table 3. Chemical composition of the pelletized concentrate ingredients (g kg⁻¹ DM).

Chemical Composition	Ground Corn	Wheat Bran	Acacia farnesiana Leaves	Sugar Cane Molasses	Soybean Meal	Dried Corn Distillers’ Grains	Cottonseed Meal
DM *	942 ± 0.89	931.7 ± 1.45	921 ± 0.88	566 ± 0.58	915 ± 0.58	915 ± 8.82	927 ± 0.33
ASH	22 ± 1.1	66.8 ± 2.38	99 ± 4.58	100.3 ± 1.15	82 ± 1.35	64.7 ± 2.40	74.1 ± 0.46
CP	73.7 ± 0.17	154.7 ± 0.14	159.7 ± 0.14	37.7 ± 0.16	463.2 ± 0.09	266 ± 0.06	399.7 ± 0.15
EE	34.9 ± 2.32	27.1 ± 0.68	17.3 ± 0.87	1 ± 0.22	14.3 ± 11.82	63.1 ± 5.17	36 ± 0.15
NDF	194 ± 0.87	481.1 ± 1.97	553.2 ± 8.11	N/D	134.5 ± 9.49	490.1 ± 2.00	417.8 ± 0.42
ADF	47 ± 1.17	167.4 ± 2.39	418.5 ± 4.07	N/D	66.6 ± 3.37	182.5 ± 0.49	184.44 ± 8.36
HEM	147 ± 2.05	313.7 ± 0.41	134.7 ± 4.05	N/D	134.5 ± 3.37	307.6 ± 2.49	233.41 ± 7.94
CEL	47.2 ± 0.65	151.6 ± 3.17	295.4 ± 4.05	N/D	66.6 ± 8.27	157.4 ± 0.37	158.09 ± 5.56
LIG	N/D	11.8 ± 1.68	117.8 ± 2.62	N/D	3.4 ± 0.07	15.5 ± 1.49	15.66 ± 1.46
TC	869.6 ± 3.38	751.5 ± 1.67	724 ± 3.75	862 ± 1.11	440.39 ± 2.18	606.3 ± 7.53	602.2 ± 21.56
NSC	675.7 ± 4.26	270.4 ± 0.31	170.7 ± 11.86	862 ± 1.11	241.1 ± 9.63	116.2 ± 5.53	72.6 ± 1.93

DM = dry matter; ASH = ash content; CP = crude protein; EE = ether extract; NDF = neutral detergent fiber; ADF = acid detergent fiber; HEM = hemicellulose; CEL = cellulose; LIG = lignin; TC = total carbohydrates; NSC = non-structural carbohydrates. N/D = not detectable value. * g kg⁻¹ NM (Natural matter).

. The Ash percentage of the leaves was 9.90%, EE was 1.73%, CP was 15.97% and NDF was 55.32%. For ground corn, the NDF content was higher than values typically reported in the literature. This discrepancy is likely due to starch interference during the NDF determination, as α-amylase was not used in the analytical procedure. Consequently, NFC values may have been underestimated for this ingredient.

3.2. Nutritional Characterization of the Experimental Treatments

Table 4. Chemical composition and in vitro dry matter digestibility of experimental diets (g kg⁻¹ DM).

	Treatment (T)
	T0	T1	T2	T3
Variable	100:0	75:25	50:50	25:75	p-Value	SEM
DM	944.5 ± 0.42 b	949.6 ± 0.13 a	942.5 ± 0.54 b	949.2 ± 1.66 a	0.0009	0.1524
ASH	40.2 ± 1.21 b	48.1 ± 0.05 a	42 ± 0.18 ab	44.3 ± 2.83 ab	0.0227	0.2495
CP	102.5 ± 6.01 bc	101.6 ± 2.74 c	112.9 ± 0.68 ab	115.7 ± 3.19 a	0.0071	0.4260
EE	38.5 ± 1.25 a	31.1 ± 1.58 b	30.3 ± 0.99 b	29 ± 0.38 b	0.0006	0.1713
NDF	405.7 ± 7.43 b	381.6 ± 6.49 b	466.9 ± 1.15 a	452.1 ± 10.31 a	<0.0001	1.2403
ADF	120.8 ± 13.73 a	112.4 ± 5.59 a	90.2 ± 16.70 a	81.8 ± 9.16 a	0.0860	1.7857
HEM	374.8 ± 53.72 a	322.8 ± 30.04 a	376.7 ± 17.84 a	370.3 ± 1.16 a	0.6079	5.5506
CEL	98.5 ± 2.16 ab	110.3 ± 6.80 a	72.4 ± 9.56 ab	66.8 ± 12.28 b	0.0200	1.4827
LIG	15.6 ± 1.46 a	15.5 ± 1.49 a	13.5 ± 1.16 a	13.9 ± 1.21 a	0.6700	0.2319
TC	763.3 ± 2.30 a	768.7 ± 4.15 a	757.1 ± 0.05 a	760.1 ± 8.06 a	0.3923	0.8099
NSC	357.6 ± 5.13 a	387.1 ± 2.34 a	290.2 ± 1.20 b	307.9 ± 40.92 b	0.0004	1.6675
ME (MJ/kg DM)	114.9 ± 0.9 a	112.5 ± 0.55 a	120.28 ± 8.06 a	110.2 ± 1.00 a	0.3963	0.7092
TPC (µg/g)	10,380 ± 534.35 b	14,250 ± 370.05 b	18,510 ± 250.04 ab	24,190 ± 270.02 a	0.0170	130.221
CT (µg/g)	2107.5 ± 53.05 d	5895 ± 106.18 c	6757.5 ± 53.07 b	7732.5 ± 53.02 a	0.0001	7.0150
IVDMD (%)	75.20 ± 0.401 a	73.84 ± 0.208 ab	73.70 ± 1.440 ab	71.12 ± 0.104 b	0.0296	1.3105

Means within a row followed by different superscript letters are significantly different (Tukey, p ≤ 0.05). Means sharing the same letter are not significantly different. DM = dry matter; ASH = ash content; CP = crude protein; EE = ether extract; NDF = detergent fiber; ADF = acid detergent fiber; HEM = hemicellulose; CEL = cellulose; LIG = lignin; TC = total carbohydrates; NSC = nonstructural carbohydrates; IVDMD = in vitro dry matter digestibility; ME = metabolizable energy; TPC = total phenolic compounds expressed in gallic acid equivalents (µg/g sample); CT = condensed tannins expressed in catechin equivalents (µg/g sample). Diet: pelleted concentrate (100:0 [T0], 75:25 [T1], 50:50 [T2], 25:75 [T3]). Results expressed as mean ± standard deviation.

shows that there was a significant difference in CP content among the treatments with the inclusion of pellets (p > 0.05). However, the variation is small; this result was expected given that the percentage of protein was adjusted during the formulation of the treatments. The variation in CP resulted from differences Although no significant difference was detected in ADF, HEM, LIG, and TC (p > 0.05), the addition of the pellets resulted in an increase in ash, CP, NDF, TPC, and CT, accompanied by a decrease in EE, CEL, NCS, and NSC. Among the variables, significant linear trends were detected for EE (p = 0.0006), NDF (p = 0.0001), NCS (p = 0.0001), TPC (p = 0.0002), CT (p < 0.0002), and CEL (p = 0.0084). Furthermore, significant quadratic trends were observed in EE (p = 0.0156) and TPC (p = 0.0122), indicating a curve-like trend.

3.2.1. Phenolic Compounds

For TPC, the calibration curve was prepared with gallic acid standards from 0 to 250 µg/mL, fitted to the equation y = 0.0046x + 0.0043 with R² = 0.999. The standard curve for tannin determination (0–300 mg/L) was described by the equation y = 0.0004x + 0.0011 with R² = 0.9885. There is a significant difference (p < 0.05) in the values of TPC and CT obtained for the different treatments, indicating an increase as the inclusion level of concentrate was higher, likely due to the increasing inclusion of Acacia farneasia leaves, which are rich in secondary metabolites [6]. Significant linear increases were observed for TPC (p = 0.0002) and CT (<0.0001), and a quadratic trend was also detected for TPC (0.0122).

3.2.2. In Vitro Digestibility of Dry Matter (IVDMD)

All treatments presented a value above 70% in IVDMD (Table 4). T0 had the highest IVDMD (75.19%); although this value was not significantly different (p > 0.05) from T1 and T2, a linear trend was detected (p = 0.0064), indicating an increase with the concentrate inclusion level.

T3 presented the lowest value (71.11%), along with the highest TPC and CT values (2419 µg/g gallic acid equivalent and 773.25 µg/g catechin equivalent). This suggests an inverse proportional relationship between secondary metabolite concentration and IVDMD, which is in accordance with previous reports [6].

With respect to ME, it is observed that no significant difference was obtained between treatments (p > 0.05). This indicates that the energy provided from each diet was similar.

3.3. In Vitro Gas Production

Table 5. Treatment effects on kinetic parameters of gas production in vitro, carbon dioxide, and methane production and ammoniacal nitrogen concentration.

	Treatment (T)
	T0	T1	T2	T3
Variable	100:0	75:25	50:50	25:75	p-Value	SEM
A, mL/g	364.43 ± 4.595 a	357.57 ± 3.493 a	352.77 ± 3.97 a	351.20 ± 4.107 a	0.1728	7.0342
kd, mL/h	0.17± 0.003 a	0.17 ± 0.005 a	0.16 ± 0.003 a	0.15 ± 0.002 a	0.0752	0.0063
L, h	2.94 ± 0.051 a	2.99 ± 0.068 a	2.88 ± 0.012 b	2.71 ± 0.013 b	0.0091	0.7595
CH4, mL/g DM	11.50 ± 1.365 a	11.60 ± 0.351 a	11.07 ± 1.041 a	9.95 ± 0.491 a	0.5833	1.5761
CO2, mL/g DM	69.23 ± 1.780 a	69.00 ± 4.5184 a	64.07 ± 2.859 a	65.25 ±3.751 a	0.5251	5.057
Ratio CH4:CO2	0.17 ± 0.0162 a	0.17 ± 0.010 a	0.17 ± 0.010 a	0.15 ± 0.020 a	0.4449	2.8158
N-NH3, mg/dL	3.48 ± 0.056 a	3.18 ± 0.034 b	2.98 ± 0.006 c	2.64 ± 0.005 d	<0.0001	0.0570

Means within a row followed by different superscript letters are significantly different (Tukey, p ≤ 0.05). Means sharing the same letter are not significantly different. A = maximum gas production from the fermentable fraction (mL), kd = constant rate of gas production (h⁻¹), L = latency time before gas production (h). Results expressed as mean ± standard deviation. Diet:pellet concentrate (100:0 [T0], 75:25 [T1], 50:50 [T2], 25:75 [T3]).

summarizes the kinetic parameters of in vitro gas production. Maximum gas production (A) was not significantly different among treatments (p > 0.05); however, a significant linear decrease was observed as the concentrate inclusion level increased (p = 0.0399), indicating a negative effect on microbiota performance; “A” values indicate that the treatment most efficiently degraded by rumen microorganisms is T0 (100:0, diet: concentrate).

“Kd” value indicates the rate of gas production (mL/h⁻¹). No significant differences were detected (p > 0.05), but the highest number was obtained for T0 and T1; in addition, a significant linear decrease (p = 0.0201) was observed.

“L” is the parameter of the latency time before gas production, for which a non-linear relation was detected, given that both linear (p = 0.0036) and quadratic (p = 0.0315) tendencies were significant. The highest value for “L” was obtained for T1, but it was not statistically different from T0, indicating a shorter colonization period of the microorganisms at higher inclusion levels of the concentrate (T2 and T3).

3.4. Methane Production, Carbon Dioxide, and Ammonia Nitrogen (N-NH₃) Concentration In Vitro

No statistical differences were observed for methane or CO₂ production with the inclusion of pelleted concentrate (p > 0.05); nevertheless, methane decreased numerically with increasing pelleted concentrate inclusion. In addition, no significant linear or quadratic trends were detected for methane, CO₂, or the CH₄:CO₂ ratio.

Table 5 also presents the effects of the treatments on the concentration of ammonia nitrogen. The calibration curve was prepared with ammonium chloride standards from 1 to 6 mM, fitted to the equation y = 0.3272x − 0.4275 with R² = 0.9825. There were significant differences (p < 0.05) among all treatments. The concentration of N-NH₃ decreased by 23.64% with increased inclusion of pelleted concentrate in the diet. A significant linear trend (p < 0.0001) was observed, indicating a strong linear decrease in this variable with the level of inclusion of the concentrate.

4. Discussion

The inclusion levels of 25, 50, and 75% (diet–pellet) were chosen to create a gradient that would allow dose-dependent effects to be detected in vitro. These levels also reflect the range that could occur in regions where A. farnesiana is locally abundant and incorporated into feeding systems. While such proportions are useful in exploring potential biological responses, their practical and economic feasibility in on-farm conditions will depend on regional resource availability and cost, which should be addressed in future in vivo and farm-level studies.

It is important to note that the use of pellets could have contributed to modifications in nutrient availability, as the densification process is known to affect protein and starch fractions [8]. While our results primarily reflect the chemical composition and fermentation characteristics of the pellets, the processing itself may partly explain some of the observed responses. This consideration highlights the relevance of evaluating A. farnesiana under conditions that closely resemble on-farm feed presentation.

The experimental diets were formulated to be isonitrogenous, ensuring that crude protein levels did not confound the observed responses. While this approach minimizes variability, it may also have reduced the visibility of certain treatment effects. It is possible that alternative balancing strategies could reveal different outcomes, and future work should explore how dietary adjustments might interact with the inclusion of A. farnesiana.

In comparison with the findings of Luna and Guerrero [20], who reported a markedly lower ash value (2.82%) than that observed in the present study (9.90%), the discrepancy may be attributed to differences in the soil conditions at the sites where the raw material was collected. Additionally, ash contents exceeding 14% are often linked to soil contamination during harvesting [21], which could further explain the variation. The ether extract (EE) content in A. farnesiana leaves (1.75%) falls within an acceptable range for use as forage. According to Arenas et al. [22], EE values above 7% can be toxic to ruminal microorganisms, potentially impairing ruminal fermentation; notably, none of the treatments in this study exceeded the threshold. Poppi et al. [23] reported crude protein (CP) contents ranging from 14% to 36% for leaves of twelve shrub species used in animal feed. The CP value obtained for A. farnesiana leaves in this study aligns well within that range, indicating its potential nutritional value. Regardless of the methodology used to assess forage quality, a feed is generally considered high quality when it contains less than 50% NDF and more than 15% CP. With a CP content of 15.96% and NDF of 55.32%, A. farnesiana leaves could be classified as medium- to high-quality forage, particularly in terms of their protein content. The unexpectedly high NDF content observed for ground corn should be interpreted with caution, as the absence of α-amylase during analysis likely caused starch to remain within the NDF fraction. This methodological limitation may have led to underestimation of NFC values, although it does not alter the overall interpretation of treatment effects.

Lignin is a key factor influencing the fiber fractions of forages [24]; as lignin concentration increases, the percentages of both ADF and NDF tend to rise. This component may contribute to the formation of lignocellulolytic complexes, which can lower digestibility and reduce gas production [25]. A. farnesiana leaves exhibited higher lignin content compared with other ingredients in the pellet formulation; nevertheless, LIG and ADF content were not significantly different among treatments, probably due to a low percentage (10%) of A. farnesiana in the pellets and to the differences in the proportion of the other components in the diets.

Furthermore, although tannins provide benefits, their anti-nutritional effects must be considered, as they can impair dietary nutritional efficiency by substantially reducing the ruminal degradation of protein and fiber, particularly hemicellulose [26]. In this study, the treatments resulted in a slight increase in HEM and a significant increase in NDF content; however, no concomitant improvements in in vitro dry matter digestibility (IVDMD) were detected. This apparent discrepancy may be partly explained by the inhibitory action of tannins on microbial fermentation, which counteracted the potential positive effects of the increased fiber availability.

As mentioned above, NDF linearly increased across treatments as the concentrate level rose; this result can be directly associated with a potential reduction in dry matter intake by the animal. It is well established that as NDF concentration rises, voluntary dry matter intake generally declines due to physical fill limitations and slower ruminal turnover [27]. Considering this, T1 is the treatment that will induce the minor effect since it was not statistically different from the control treatment (T0). The current study did not include measurements of fiber degradation kinetics (e.g., NDF and ADF disappearance), which would have provided a more direct link between lignin and tannin concentrations and digestibility outcomes. This remains an important avenue for future research to better understand how A. farnesiana affects fiber utilization in ruminants. Furthermore, T1 exhibited the highest proportion of nonstructural carbohydrates (NS C), which could contribute to lowering methane synthesis by shifting ruminal fermentation towards a lower acetate: propionate ratio—a change often linked to reduced hydrogen availability for methanogenesis. NSC may also have contributed to a higher IVDMD as it is constituted of easily fermentable carbohydrates.

The slight variation in CP content among treatments (Table 4) may be attributed to differences in ingredient composition, as well as seasonal fluctuations in the nutritional profile of the forages used [28]. Importantly, all CP values reported here are within acceptable limits, considering that ruminant cattle require 8–10% CP in their diet to ensure adequate intake, rumen function, and growth [29].

Treatments T1 and T2 contained approximately 6% of CT on a dry matter basis (Table 4). Piñeiro-Vázquez et al. [30] reported that incorporating CT into ruminant diets at levels between 3–6% of DM can reduce methane emissions while also improving weight gain and milk production; however, although smaller values were obtained as the inclusion concentrate level increased, no statistically significant reductions were observed in this study. It has been reported that moderated CT concentrations positively influence protein metabolism in ruminants by decreasing protein degradation in the rumen and enhancing amino acid absorption in the small intestine [31], an effect that must be explored in future approaches, as well as the compositional changes in function of the season. In this respect, Zárate-Martínez et al. [32], noted that CT synthesis in plants increases under biotic or abiotic stress conditions; this suggests that the dry season may enhance tannin concentration in A. farnesiana leaves, amplifying their nutritional and functional effects. Furthermore, regional origin and plant part studied can significantly affect the concentrations of CT and phenolic compounds (PC) in Acacia species [25,33], which may explain differences between the present results and those reported by Torres-Velázquez et al. [34], who analyzed A. farnesiana pods rather than leaves.

Regarding metabolizable energy (ME), no significant differences were detected among treatments (p > 0.05); this is expected because the treatments were formulated as balanced diets using the Oklahoma State University beef ration calculator software that considers the chemical composition of the ingredients. Nevertheless, T3 (50:50 inclusion) exhibited the highest ME value (11.55 Mcal/kg DM), with an overall composition consistent with that described as acceptable by Piñeiro-Vázquez et al. [30]. Rubanza et al. [35], studying the effect of Acacia leaf CT on in vitro digestibility with polyethylene glycol, found that ME increased as CT concentration decreased. In addition, CTs have documented antimicrobial and antifungal properties, and excessively high concentrations can inhibit ruminal fermentation [33,36]. However, the CT concentrations in this study remain within a safe and functional range, as no adverse effects on ME were observed.

Di Marco [37] emphasizes that forages of good quality are generally characterized by an in vitro dry matter digestibility (IVDMD) greater than 70%. Based on this criterion, all treatments in the present study can be classified as good-quality diets, since all exceeded the 70% threshold (Table 5). The highest IVDMD value was recorded in T0 (75.2%), which can be attributed to its lower forage proportion [38] and the absence of A. farnesiana pellets in its composition. This observation is consistent with previous reports indicating that high concentrations of CT can reduce in vitro digestibility [34]. For instance, T3, which contained the highest CT concentration (773.25 μg/g catechin equivalent), exhibited the lowest IVDMD (71.11%).

At the measured levels, condensed tannins showed both promising and potentially limiting effects. On one hand, their inclusion appeared to mitigate methane production and reduce ammonia nitrogen concentrations, which aligns with their known capacity to modulate ruminal fermentation. On the other hand, condensed tannins may also reduce digestibility and limit nutrient availability. These trade-offs highlight the need for cautious interpretation of the results and underline that the benefits of methane reduction must always be considered in light of possible nutritional drawbacks.

Regarding maximum gas production (“A”), no statistical differences were observed. Small reductions are likely related to the variation in CT concentration among treatments, which did not appear to negatively impact ruminal fermentation [39].

Lag time (L) differed significantly between treatments (p < 0.05), and a strong linear reduction was observed (p = 0.0036), indicating that ruminal microorganisms are colonizing the substrate before gas production. This reduction may be due to the higher diversity of components present in the concentrate to which the rumen microbiota must adapt in order to degrade it [17].

Fermentation rate (Kd, h⁻¹) did not significantly change among treatments p > 0.05), indicating that this parameter is not affected by the inclusion of the concentrate. Abdulrazak et al. [40] reported lower fermentation rates for A. nubica (0.1165 h⁻¹) and A. brevispica (0.0295 h⁻¹) than those observed in this study; nevertheless, direct comparisons can not be made as A. farnesiana was not evaluated alone in this study.

The highest gas production was recorded for the control (T0), but it was not significantly different from the other treatments; this behavior was observed for CO_2, which is expected as carbon dioxide is the main component of gas produced in the rumen. Velázquez et al. [39] reported that inclusion of 200 g/kg of A. farnesiana reduced IVDMD and gas production. Higher inclusion rates, such as 300 g/kg, further reduced gas production, potentially affecting microbial protein synthesis in the rumen [39]. In the present study, the higher pellet inclusion levels in T2 and T3 (500 and 750 g/kg, respectively) corresponded to 50 g/kg and 75 g/kg of A. farnesiana in the diets, which is not enough to contribute to reducing gas production. Slightly lower methane production observed in diets with higher concentrate proportions may be explained by faster fermentation rates, as more digestible diets tend to produce less methane [41]. Additionally, the inclusion of secondary metabolites naturally present in A. farnesiana leaves could have contributed to this effect [42]. Although methane production showed a tendency to decrease with increasing levels of A. farnesiana, the differences were not statistically significant. This suggests that dose–response effects may exist but require confirmation under experimental conditions with greater replication or more sensitive measurement techniques. Further studies are needed to verify these trends and to determine the threshold levels at which tannins consistently influence methane emissions. The effects of treatments on ammonia nitrogen (N–NH₃) concentration, a key indicator of in vitro ruminal fermentation, are shown in Table 5. N–NH₃ concentration decreased by 23.64% with increasing inclusion of pelleted concentrate, a strong linear reduction was also detected. According to Araiza-Rosales et al. [43], such changes reflect ongoing proteolysis, which may increase when crude protein levels are elevated through the inclusion of A. farnesiana pellets. Some studies have suggested that increases in N–NH₃ concentration indicate that dietary protein is not being efficiently incorporated into microbial protein synthesis, resulting in energy loss for the ruminant [43]. In this study, T0 showed the highest N–NH₃ concentration (3.47 mg/dL), but none of the treatments reached the threshold for optimal microbial protein synthesis, which Rodríguez et al. [44] set at >5 mg/dL. Pulido F. [45] notes that the optimum range for microbial growth is 3–9 mM/L N–NH₃, equivalent to approximately 8 mg/dL. The relatively low values observed here may be related to alterations in ruminal microbial dynamics and processes [46], suggesting that increasing crude protein inclusion in the diet formulation could be necessary to optimize microbial activity and protein synthesis.

5. Conclusions

The findings of this study indicate that incorporating a pelleted concentrate containing 10% A. farnesiana into a ruminant diet at a 25% inclusion rate did not alter most nutritional parameters compared to the control. Notably, the inclusion of this pellet markedly increased the concentrations of total phenolic compounds and condensed tannins. This increase in secondary metabolites is the most plausible explanation for the observed reductions in in vitro dry matter digestibility and ammoniacal nitrogen (N-NH₃). The results suggest that A. farnesiana could serve as a viable feed alternative for cattle, as it did not increase methane production and may act as a ruminal fermentation modulator. However, its potential negative effects on digestibility must be considered. Therefore, the optimal inclusion threshold for A. farnesiana must be carefully determined in future studies. Furthermore, comprehensive in vivo trials and mechanistic investigations are essential to validate and expand upon these in vitro observations.