Evaluation of Nutritional Value and Rumen Degradation Rate of Six Unconventional Feeds Using In Vitro and In Situ Methods

Abstract

Objective: This study systematically evaluated the nutritional compositions and bioactive compounds of six unconventional feed resources (Pepper residue (PR), Grape marc (MC), Pepper straw (PS), Lycium barbarum branches and leaves (LBBL), Licorice straw (LS), and Cyperus esculentus leaves (CES)). It also assessed the rumen degradability and rumen fermentation characteristics at different substitution levels through in vitro and in situ methods, to explore their potential application in sheep diets. Methods: Samples were analyzed considering nutrient composition, amino acids, polyunsaturated fatty acids (PUFAs), and bioactive compounds. In situ degradation was measured using rumen-fistulated sheep, and in vitro batch fermentation culture was conducted at varying substitution levels (0–100%) to measure gas production, pH, VFAs, NH₃-N, and microbial crude protein (MCP). Results: The six unconventional feed resources showed significant differences in nutrient composition, bioactive compounds, and fermentation performance. Crude protein (CP) ranged from 4.45% to 15.76%, with LS highest in total amino acids. LBBL contained 4.24 g/kg Lycium barbarum polysaccharides, LS had 9.24 g/kg liquiritin, GM was richest in proanthocyanidins, and PS had more capsaicin than PR. PR exhibited the highest DM degradation (74.77%, p < 0.001), followed by LS; CEL was lowest. PR and LS also had the highest CP degradation. In vitro fermentation revealed significant differences in fermentation characteristics among the six feeds. At 100% replacement, PR and LS exhibited high cumulative gas production, elevated MCP concentrations, and total VFAs of 54.41 and 64.02 mmol/L (p < 0.001), respectively. At 25% replacement, GM and CEL achieved high concentrations of VFAs and maintained MCP levels of 27.84 and 31.57 mg/dL (p < 0.001). PS reached its maximum total VFAs and MCP at 50% replacement, while LBBL reached 64.90 mmol/L total VFAs and 32.63 mg/dL MCP at 75% replacement. Conclusions: Nutrient composition and degradation kinetics varied significantly among substrates. PR had the highest DM degradability, while CEL had the lowest. PR and LS maintained stable fermentation at 100% substitution. GM and CEL were most effective at 25%; PS at 50%; and LBBL at 75% substitution levels.

1. Introduction

Traditional livestock farms account for approximately 26% of the global total area, and contribute approximately 60% of the output of animal products [1]. However, as global population numbers rise and demand for livestock products increases, land use conflicts between humans and livestock are becoming increasingly severe. Meanwhile, growing concerns about health have led to stricter regulations on the use of medications in livestock production. Therefore, the development and utilization of forage crops containing high levels of bioactive compounds play a key role in animal nutrition and livestock production. Non-conventional forage refers to low-cost plant-based raw materials with uncertain nutritional value that are rich in plant secondary metabolites and seldom used in animal production, including crop residues, plant processing by-products, forestry by-products, and dregs.

Non-conventional feed materials such as pepper by-products, grape marc, Lycium barbarum branches and leaves, Licorice straw, and Cyperus esculentus leaves have potential in animal nutrition due to being rich in bioactive compounds. Pepper straw and residue contain capsaicin (CAP), which possesses antioxidant, anti-inflammatory, and antimicrobial properties. Studies have shown that CAP improves average daily gain and gut health in weaned piglets, increases feed intake in lactating cows, and enhances milk yield and serum metabolites in buffaloes [2,3]. In dairy cows, CAP supplementation increased dry matter intake, and high-dose CAP (4 mg/kg DM) improved milk production, milk composition, and serum metabolites [2]. In weaned piglets, 2% dietary capsaicin significantly improved average daily gain, antioxidant capacity, and digestive enzyme activity, while reducing intestinal inflammation [3]. Grape marc (GM), a by-product of winemaking, is rich in polyphenols with antioxidant, antimicrobial, and antitumor activities. Including GM in lamb diets can enhance antioxidant capacity, regulate gut microbiota, promote growth performance, and improve meat quality. Specifically, supplementing sheep diets with 5% grape marc had a positive impact on carcass traits and meat quality [4]. Lycium barbarum’s main bioactive compound is Lycium barbarum polysaccharide (LBP), which can elevate immune factors, improve growth and immune function in piglets and broilers, and modulate gut microbiota [5,6]. In lambs, a mixture of LBPs and Astragalus membranaceus polysaccharides improved rumen fermentation and growth [7]. For example, adding 2000 mg/kg LBPs to broiler diets can significantly improve body weight, daily weight gain, feed intake, and feed conversion ratio [5]. Licorice (Glycyrrhiza glabra), rich in saponins and flavonoids, exhibits immunoregulatory and antioxidant effects. Supplementation with licorice extract has been shown to enhance antioxidant capacity in sheep meat, improve immune status in calves, and promote growth and intestinal health in beef cattle [8,9]. Specifically, 3000–4000 mg/kg licorice extract enhanced meat antioxidant capacity in Tan sheep [8]; 10 g/day of licorice root extract increased serum immunoglobulins in Holstein calves [9]. Cyperus esculentus, due to its nutritional content and bioactive ingredients, can support intestinal barrier function and stimulate probiotic growth [10]. For example, replacing 33.3% of maize in cockerel diets improved carcass yield and reduced feed cost [11]. In goats and poultry, inclusion of tiger nut improved protein and fiber digestibility, carcass yield, and showed potential for liver protection and blood glucose reduction [11,12].

Different types of feed and their usage ratios can affect the rumen degradation and in vitro gas production in sheep. This study evaluated the feed characteristics of six unconventional feed resources using in vitro fermentation and in situ rumen degradation methods and determined their appropriate substitution levels in sheep diets.

2. Materials and Methods

2.1. Material

2.1.1. Ethics Committee Approval

The sheep were purchased from Tumushuke Anxin Animal Husbandry, Tumushuke, China. The experimental procedures were conducted in accordance with the ethics and animal welfare committee of the Universidad de Tarim (approval protocol PA20250326010). Rumen cannulation (inner diameter 4 cm) was surgically performed at the Tarim University Veterinary Hospital under Animal Welfare Committee approval. The cannula site was cleaned weekly. During the in situ incubation, 3–5 nylon bags were inserted simultaneously.

2.1.2. Raw Materials for Experiments

The experimental materials used in this study were sampled in typical planting areas in the Xinjiang Region (Yanqi County, Awati County, Alar City) in August 2024. The collected materials included six types of non-traditional roughage resources: pepper straw, Pepper residue, grape marc, Licorice straw, Lycium barbarum branches and leaves, and Cyperus esculentus leaves. The whole process followed the ISO 6497:2002 [13] feed sampling standard, and at least three parallel samples were prepared to ensure the statistical validity of the analytical data. The collected samples were air dried in a cool and ventilated place (temperature 25 ± 3 °C, relative humidity < 40%) and were regularly turned over until reaching a constant weight (moisture content ≤ 10%). Then, the feed was pulverized with a high-speed pulverizer, passed through a 0.425 mm sieve, and sealed for preservation for subsequent analyses.

2.1.3. Experimental Animal and Feeding Management

Twelve Hu sheep with permanent rumen fistulas (30.0 ± 2.8 kg, 12 months of age) were purchased in Tumushuke, China. The experimental sheep were housed individually in separate pens and managed separately (1.5 m × 1.2 m, with a slatted floor and individual feeders). The dietary feed was formulated according to the guidelines for meeting the nutrient requirements of sheep and allowing them to gain 300 g per day (Nutrient Requirements of Meat Sheep [14]). Feed was provided daily at 09:00 and 18:00. All sheep had ad libitum access to clean water. The diet was formulated at a 4:6 concentrate-to-roughage ratio (

Table 1. Ingredients and nutrient levels of the experimental diet (dry matter basis, %).

Ingredients	Content	Nutrients Levels †	Content
Alfalfa	9.00	Metabolic energy ‡ (MJ/kg)	9.99
Whole plant corn silage	5.50	CP	15.72
Corn straw	45.50	NDF	38.47
Corn	12.00	ADF	21.26
Wheat bran	7.00	EE	6.26
Soybean meal	4.00	NFC	22.41
Cottonseed meal	12.70	NFC/NDF	0.58
CaHPO4·2H2O	1.30	Ca	0.98
NaCl	1.00	TP	0.66
Premix *	2.00	Ash	10.21

* Premix provided the following per kg in basic diets: 10 mg of iron, 135 mg of manganese, 100 mg of zinc, 0.5 mg of cobalt, 12.5 mg of copper, 0.3 mg of selenium, 1.5 mg of iodine, 1400 IU of vitamin A, 500 IU of vitamin D, and 50 mg of vitamin E. † Nutrients were determined on a dry matter basis at 105 °C. ME, metabolic energy; CP, crude protein; NDF, neutral detergent fiber; ADF, acid detergent fiber; NFC, non-fibrous carbohydrate; TP, total phosphorus. ‡ The metabolic energy and non-fibrous carbohydrates in the nutritional level were calculated about the Nutritional Requirements of Sheep for Meat in China, ME = 0.046 + 0.820 × (17.211 − 0.135 × NDF), and the rest were measured values [14].

).

2.2. Experimental Design

2.2.1. In Situ Nutrient Digestibility

The disappearance of DM and CP was determined according to AOAC methods 942.05 and 984.13 [15], respectively, while NDF and ADF were analyzed following the standardized procedures of Van Soest [16]. The feed ingredients tested were PR: Pepper residue, GM: Grape marc, PS: Pepper straw, LBBL: Lycium barbarum branches and leaves, LS: Licorice straw, and CEL: Cyperus esculentus leaves. In short, approximately 3 g of sample (on an air-dry matter basis) was weighed and placed into nylon bags (4.5 × 10 cm, 40 µm pore size) labeled with numerical codes. The bags were then incubated in the rumen for 0, 6, 12, 24, 36, 48, and 72 h before being immediately removed. In the in situ trial, each sheep was considered an experimental unit. This study used a 6 × 6 Latin square design with repeated randomization, consisting of six observation periods using twelve sheep and six different forages. At the end of the experiment, all bags (including 0 h) were placed in a bucket filled with cold tap water to halt microbial fermentation. After manual cleaning in cold tap water, the bags were dried in an oven at 65 °C for 48 h. The extracted material was weighed using an analytical balance. Subsequently, the residual material from the nylon bag was ground, passed through a 0.425 mm sieve, and prepared for nutrient analysis. Residues from the bags were pooled according to incubation time and treatment.

2.2.2. In Vitro Fermentation

On the basis of DM, the six aforementioned materials were used to replace corn straw in the basal diet at substitution levels of 0%, 25%, 50%, 75%, and 100%. The prepared formula feed was crushed using a grinder (KX-500A, Beijing Zhongxing Ltd., Beijing, China) and then passed through a 0.425 mm sieve. All animals were adapted to the experimental diet for 15 days prior to sampling. The rumen contents are a mixed sample from twelve sheep. Before morning feeding, rumen fluid was collected from twelve Hu sheep fitted with permanent fistulas, squeezed through a four-layer cheesecloth to remove feed particles, preserved under anaerobic conditions at 39 °C, and immediately transported to the laboratory. The in vitro gas production method is described in Menke [17]. Fermentation was carried out using a DSHZ-300A (Taicang Experimental Equipment Co., Jiangsu, China) water bath thermostat shaker device, weighing about 200 mg of the sample, using a long strip of paper placed into a glass syringe coated with petroleum jelly for the in vitro gas production test, with 3 replicates per sample and 3 blank tubes. The artificial rumen fluid was prepared according to Menke et al. [17] (

Table 2. Chemical compositions of the buffer (according to Menke et al., 1979 [17]).

Ingredient Composition	Amount (g/L DH2O)
NaHCO3 (Sodium carbonate, Solution B)	39.00
Na2HPO4 (Sodium phosphate, Solution C)	5.70
KH2PO4 (Potassium dihydrogen phosphate, Solution C)	6.20
MgSO4·7H2O (Magnesium sulfate heptahydrate, Solution C)	0.60
CaCl2·2H2O (Calcium chloride dihydrate, Solution A)	0.13
MnCl2·4H2O (Manganese(II) chloride tetrahydrate, Solution A)	0.10
CoCl2·6H2O (Cobalt(II) chloride hexahydrate, Solution A)	0.01
FeCl3·6H2O (Ferric chloride hexahydrate, Solution A)	0.08
Resazurine (0.1%, w/v)	1.00 mL
Reduction solution (95 mL H2O, 4 mL 1 N NaOH, 625 mg Na2S·9H2O)	40.00 mL

) by sequentially mixing 400 mL of distilled water, 0.1 mL of solution A, 200 mL of solution B, 200 mL of solution C, 1 mL of resazurine (0.1%, w/v), and 40 mL of the reduction solution. The prepared buffer was then mixed with rumen liquor at a ratio of 2:1 (buffer:rumen liquor, v/v) to obtain the incubation medium. CO₂ was introduced, and 30 mL of the culture solution was added to the culture tube using a dispenser to remove the air bubbles, and the culture tube was placed in a 39 °C constant-temperature shaking incubator for cultivation and timing. When cultured for 0, 6, 9, 12, 24, 48, and 72 h, we took the culture tube to quickly read the scale value of the piston and recorded it. All in vitro incubations were conducted in three independent batches, and each incubation bottle was regarded as an experimental unit. After 72 h incubation, the culture tube was rapidly transferred to an ice-water bath to terminate fermentation. The fermented broth was filtered through four layers of gauze, aliquoted, and its pH was measured before storage at −80 °C for subsequent analyses.

2.3. Measurement of Treatments and Methods

2.3.1. Determination Method of Nutrient Content

In this experiment, the determination of feed nutrient content was conducted following the 22nd edition of the AOAC International Official Methods of Analysis [15]. Dry matter (DM) was determined by oven-drying at 135 °C for 2 h (930.15), the determination of crude ash content is carried out by incineration at 550 °C ± 15 °C until a constant weight is achieved (942.05), crude protein (CP) by Kjeldahl nitrogen determination (N × 6.25; 984.13), ether extract (EE) using Soxhlet extraction with diethyl ether (920.39), NDF and ADF via sequential detergent fiber analysis with α-amylase (Van Soest method [16]), calcium (Ca) by atomic absorption spectroscopy (968.08), phosphorus (P) via vanadate-molybdate colorimetry (965.17), vitamins by HPLC (Agilent 1260 Infinity II, Agilent Technologies, Waldbronn, Germany) with C18 column (992.03), and amino acids through acid hydrolysis (6 M HCl, 110 °C, 24 h) followed by HPLC (Agilent 1260 Infinity II) derivatization (994.12).

2.3.2. Bioactive Ingredient Content

Capsaicin, gallocatechin, and liquiritin contents were quantified using high-performance liquid chromatography (HPLC) with reference to AOAC Official Methods. Lycium barbarum polysaccharides were determined according to the Pharmacopoeia of the People’s Republic of China (Volume I, 2020 edition). Proanthocyanidin content was analyzed following the DB12/T 885-2019 [18] standard (Tianjin local standard, China), validated by the Folin–Ciocalteu colorimetric method.

2.3.3. Rumen Fermentation Parameters

pH was measured using a Testo205 pH meter (Testo SE & Co., KGaA, Lenzkirch, Germany); ammonia-N concentration was determined by the phenol-hypochlorite colorimetric method [19] after centrifugation (3500–4000 rpm, 10 min); microbial crude protein (MCP) was quantified via Coomassie Brilliant Blue G-250 [20]; and volatile fatty acids (VFAs) were analyzed using an Agilent 6890N gas chromatograph with an HP19091N-213 capillary column (Agilent Technologies, Santa Clara, CA, USA; injector: 200 °C, N₂ carrier gas: 2.0 mL/min) [21].

2.3.4. Analysis of Selected Data on In Situ Rumen Degradation Rates

The results of DM and CP disappearance from nylon bags were fitted into the following exponential equation of Ørskov [22] using non-linear regression (SPSS 26):

P = a + b(1 − e^(−ct))

where P is the disappearance of nutrients during time t, a is the soluble nutrient fraction rapidly washed out of the bags and assumed to be completely degradable, b is the proportion of insoluble nutrients potentially degradable by microorganisms, e is the natural logarithm, c is the degradation rate of fraction b per hour (i.e., k) and t is a time of incubation.

The effective degradability (ED) of DM, CP, NDF, and ADF in situ of each feed sample for each of the three lambs was calculated as follows:

ED = a + (b × c)/(c + k)

where ED denotes the effective degradation rate (%); k is the estimated rate of outflow from the rumen and a, b, and c are the same parameters as described above.

2.4. Statistical Analyses

The data set was organized using Excel 2010, and a single-factor analysis of variance (ANOVA) was conducted with SPSS 26.0 for statistical analysis. Additionally, all graphical representations were prepared using GraphPad Prism (version 10.0).

Data were analyzed according to the following model:

$$Y_{ij}=\mu +T_i+e_{ij}$$

where $Y_{ij}$ is the observed value, $\mu$ is the overall mean, $T_i$ is the fixed effect of treatment (feed type), and $e_{ij}$ is the random residual error. The normality of residuals and the homogeneity of variances were assessed by inspecting residual plots. Outliers and influential observations were checked using studentized residuals, and values exceeding ±3 standard deviations from the mean were excluded from the analysis.

The results from the tests were expressed as the mean and the standard error of the mean (SEM). Duncan’s multiple range test was used to compare statistical differences between the treatments, with p ≤ 0.05 indicating significance, and 0.05 < p ≤ 0.10 considered a trend.

3. Results

3.1. Nutritional Composition of Experimental Raw Materials

3.1.1. Nutrient Levels

As shown in

Table 3. Chemical composition of six raw materials (%, on dry matter basis, mean ± SEM, n = 3).

Items	Nutrient Levels
	PR	GM	PS	LBBL	LS	CEL
DM, %	88.11 ± 0.07	89.64 ± 0.01	92.80 ± 0.10	93.29 ± 0.04	90.30 ± 0.05	89.76 ± 0.02
Ash, %	12.42 ± 0.18	7.40 ± 0.17	11.67 ± 0.14	9.17 ± 0.22	7.63 ± 0.15	16.42 ± 0.97
CP, %	15.76 ± 0.10	10.84 ± 0.19	14.61 ± 0.29	14.93 ± 0.30	14.00 ± 0.11	4.45 ± 0.03
EE, %	1.91 ± 0.20	6.53 ± 0.07	1.33 ± 0.31	5.88 ± 1.09	4.53 ± 0.14	6.47 ± 0.06
NDF, %	35.71 ± 0.62	55.36 ± 1.69	48.96 ± 2.58	41.17 ± 0.39	58.01 ± 1.28	69.50 ± 0.56
ADF, %	30.35 ± 0.52	50.42 ± 1.69	31.56 ± 2.25	31.52 ± 0.16	46.78 ± 1.27	48.73 ± 1.43
Ca, %	0.66 ± 0.02	0.67 ± 0.01	0.97 ± 0.22	1.26 ± 0.14	1.73 ± 0.33	1.13 ± 0.27
P, %	0.48 ± 0.02	0.31 ± 0.13	0.14 ± 0.01	0.22 ± 0.01	0.16 ± 0.01	0.18 ± 0.02
VA, ng/g	0.06 ± 0.01	0.70 ± 0.01	0.74 ± 0.01	0.32 ± 0.01	0.70 ± 0.01	0.37 ± 0.01
VC, μg/g	1262.33 ± 22.40	342.33 ± 7.54	1373.00 ± 19.97	1529.67 ± 6.49	1121.67 ± 13.78	760.00 ± 10.79
VE, μg/g	30.93 ± 0.47	113.67 ± 2.33	210.00 ± 5.69	55.97 ± 1.54	143.33 ± 2.19	186.33 ± 4.33
VB1, mg/g	16.97 ± 0.13	2.62 ± 0.01	6.82 ± 0.08	10.02 ± 0.04	9.03 ± 0.10	4.24 ± 0.07
VB2, ng/g	0.51 ± 0.01	1.02 ± 0.01	0.52 ± 0.01	0.83 ± 0.01	0.64 ± 0.01	0.96 ± 0.01

PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: Licorice straw; CEL: Cyperus esculentus leaves. DM: dry matter; Ash: crude ash; CP: crude protein; EE: ether extract; NDF: neutral detergent fiber; ADF: acid detergent fiber; Ca: calcium; P: phosphorus; VA: vitamin A; VC: vitamin C; VE: vitamin E; VB1: vitamin B1; VB2: vitamin B2. All values are presented on a dry matter basis unless otherwise indicated.

, DM content ranged from 88.11 to 93.29%. CP content was highest in pepper residue and lowest in CEL. NDF levels ranged from 48.96% to 69.50%, while ADF values varied between 31.56% and 50.42%. As indicated by the findings of the vitamin content determination, the materials under scrutiny have been shown to be abundant in Vitamin C, with the highest recorded concentrations found in the LBBL, reaching 1529.67 μg/g. On the other hand, the vitamin A content was minimal, with no individual value exceeding 0.74 ng/g.

3.1.2. Amino Acid Content

The highest concentrations of threonine, leucine, isoleucine, phenylalanine, valine, and total essential amino acids were found in LS among the six materials (

Table 4. Amino acid content of the six raw materials (dry matter basis, g/100 g protein, mean ± SEM, n = 3).

Items		Amino Acid Content
		PR	GM	PS	LBBL	LS	CEL
Essential amino acids (EAAs)	Threonine	0.43	0.44	0.52	0.49	0.54	0.13
	Lysine	0.37	0.52	0.76	0.89	0.73	0.20
	Leucine	0.67	0.66	0.83	0.80	0.99	0.26
	Isoleucine	0.36	0.40	0.44	0.45	0.52	0.14
	Methionine	0.05	0.04	0.04	0.01	0.03	0.01
	Phenylalanine	0.40	0.38	0.55	0.47	0.57	0.14
	Valine	0.43	0.46	0.55	0.55	0.61	0.17
Total EAAs		2.69	2.89	3.68	3.64	3.99	1.04
Non-essential amino acids (NEAAs)	Alanine	0.47	0.44	0.58	0.54	0.60	0.25
	Proline	1.32	0.70	1.17	1.42	1.13	0.18
	Cystine	0.09	0.05	0.03	0.03	0.04	<0.01
	Tyrosine	0.24	0.23	0.31	0.28	0.44	0.08
	Aspartic Acid	1.58	0.50	0.56	0.72	0.58	0.13
	Glutamic Acid	1.19	1.13	1.00	1.30	0.71	0.21
	Serine	0.45	0.43	0.48	0.52	0.49	0.13
	Glycine	0.56	0.69	0.60	0.56	0.66	0.17
	Histidine	0.16	0.32	0.25	0.25	0.26	0.05
	Arginine	0.51	0.58	0.56	0.51	0.66	0.12
Total NEAAs		6.58	5.07	5.54	6.12	5.57	1.33
Total AAs		9.27	7.96	9.23	9.77	9.56	2.37

PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: Licorice straw; CEL: Cyperus esculentus leaves.

). The maximum lysine concentration was observed in PS. With the exception of methionine, which had the lowest concentration in LBBL, CEL consistently exhibited the lowest levels (e.g., total essential amino acids 1.04 g/100 g, compared to 3.99 g/100 g in LS) of the other six EAAs, total EAAs, and total AAs.

3.1.3. Polyunsaturated Fatty Acids Content

As shown in

Table 5. Polyunsaturated Fatty Acid Content of Six Raw Materials (dry matter basis, g/100 g).

Items	Polyunsaturated Fatty Acids Content
	PR	GM	PS	LBBL	LS	CEL
cis,cis-9,12-octadecadienoic acid(C18:2n6c)	0.05	1.64	0.22	0.14	0.07	0.04
cis,cis,cis-9,12,15-Octadecatrienoic acid(C18:3n3)	0.05	0.09	0.36	0.11	0.43	0.05
cis-9-octadecenoic acid(C18:1n9c)	0.01	0.34	0.03	0.01	0.02	0.04
Dodecanoic acid(C12:0)	<0.01	<0.01	0.00	<0.01	0.01	<0.01
Tetradecanoic acid(C14:0)	0.00	<0.01	0.01	<0.01	0.01	<0.01
Hexadecanoic acid(C16:0)	0.02	<0.01	0.14	0.06	0.09	0.02
Heptadecanoic acid(C17:0)	<0.01	<0.01	0.00	<0.01	<0.01	<0.01
Octadecanoic acid(C18:0)	<0.01	0.07	0.03	0.01	0.00	<0.01
Eicosanoic acid(C20:0)	<0.01	0.01	0.01	0.01	0.00	<0.01
Docosanoic acid(C22:0)	<0.01	0.01	0.01	0.00	0.03	<0.01
Tetracosanoic acid(C24:0)	<0.01	0.01	0.01	<0.01	0.02	<0.01

PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: Licorice straw; CEL: Cyperus esculentus leaves.

, in the analysis of polyunsaturated fatty acid (PUFA) content, the highest concentrations of C18:2n6c, C18:1n9c, C18:0, and C20:0 were found in GM. LS was characterized by maximal levels of C18:3n3, C12:0, C22:0, and C24:0, while PS exhibited the highest values for C14:0, C16:0, and C17:0. Minimal PUFA contents were identified as follows: C18:3n3 in PR, C16:0 in GM, C18:1n9c in LBBL, and C18:2n6c in CEL. Additionally, trace levels of other PUFAs were found in CEL.

3.1.4. Bioactive Ingredient Content

As shown in

Table 6. Bioactive ingredient content of Six Raw Materials (dry matter basis, g/kg).

Items	Capsaicin (PR)	Proanthocyanidins (GM)	Capsaicin (PS)	Lycium Barbarum Polysaccharides (LBBL)	Liquiritin (LS)	Epigallocatechin (CEL)
Content	0.02	8.48	0.08	9.24	4.24	1.77

PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: Licorice straw; CEL: Cyperus esculentus leaves.

, capsaicin levels were detected at extremely low concentrations in pepper residue (<0.02 g/kg), whereas a significantly higher capsaicin content was quantified in pepper straw (0.08 g/kg). Liquiritin, a major flavonoid in Licorice straw, exhibited a notably high concentration of 9.24 g/kg. The polysaccharides derived from Lycium barbarum branches and leaves were identified as the most abundant bioactive compound, reaching 4.24 g/kg. In contrast, epigallocatechin, isolated from Cyperus esculentus leaves, was present at a relatively low concentration of 1.77 g/kg. Proanthocyanidins in grape marc were quantified at 8.48 g/kg.

3.2. In Situ Nutrient Digestibility and Degradation Parameters

As demonstrated in Figure 1, the ruminal degradation rates of DM, CP, NDF, and ADF for all six materials were observed to progressively increase over time. In Supplementary Table S1, the 72 h degradation rates of DM, CP, NDF, and ADF in PR were significantly higher than those of other materials (p < 0.001), whereas CEL had significantly lower DM and CP degradation rates (p < 0.001). Additionally, LS had significantly lower NDF and ADF degradation rates compared to the other materials (p < 0.001).

Significant differences were observed among the six raw materials regarding ruminal disappearance and fermentation parameters for DM, CP, NDF, and ADF (

Table 7. Rumen disappearance fermentation parameters (dry matter basis, %, n = 3).

Items		Group						SEM	p-Value
		PR	GM	PS	LBBL	LS	CEL
DM	a/%	46.29 a	22.81 c	15.67 d	9.15 e	36.27 b	2.45 f	3.74	<0.001
	b/%	52.32 bc	73.16 a	78.05 a	34.69 d	57.91 b	46.57 c	3.69	<0.001
	c/(%/h)	0.04 ab	0.01 c	0.02 bc	0.04 ab	0.04 ab	0.04 a	0.00	0.022
	ED/%	74.77 a	45.14 c	45.23 c	27.56 e	66.51 b	29.82 d	4.23	<0.001
CP	a/%	3.90 b	2.94 bc	2.40 cd	1.28 de	6.41 a	0.38 e	0.49	<0.001
	b/%	11.46 b	11.81 b	17.915 a	5.93 b	13.30 b	4.36 b	5.31	<0.001
	c/(% /h)	0.10 a	0.01 bc	0.00 c	0.03 b	0.03 b	0.02 bc	0.01	<0.001
	ED/%	12.47 a	6.00 b	4.95 c	4.15 d	12.12 a	1.86 e	0.97	<0.001
NDF	a/%	24.88 a	12.22 c	9.45 d	2.96 e	21.77 b	2.22 e	2.10	<0.001
	b/%	68.43 a	55.05 b	69.69 a	23.71 d	54.61 b	38.770 c	4.09	<0.001
	c/(% /h)	0.03 a	0.02 b	0.02 b	0.02 b	0.04 a	0.04 a	0.00	<0.001
	ED/%	59.99 a	29.45 d	32.17 c	10.89 f	50.58 b	23.46 e	3.56	<0.001
ADF	a/%	23.13 a	11.07 c	7.66 d	2.41 e	15.77 b	1.53 e	1.84	<0.001
	b/%	69.73 a	45.99 b	65.72 a	17.91 c	37.28 b	24.19 c	4.82	<0.001
	c/(%/h)	0.02 b	0.02 bc	0.01 c	0.02 bc	0.04 a	0.03 a	0.00	<0.001
	ED/%	51.27 a	26.17 c	25.74 c	8.74 e	35.78 b	13.83 d	3.39	<0.001

DM: dry matter; CP: crude protein; NDF: neutral detergent fiber; ADF: acid detergent fiber; a: rapidly degradable fraction (%); b: potentially degradable fraction (%); c: degradation rate constant of fraction b (%/h); ED: effective degradability (%), calculated at a rumen outflow rate of 0.08/h. Different superscript letters indicate significant differences among treatments (p ≤ 0.05). PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: Licorice straw; CEL: Cyperus esculentus leaves.

), with all parameters demonstrating p < 0.001 except for the degradation rate of DM (p = 0.022). For DM, the rapidly degradable fraction and effective degradability of PR were found to be significantly higher than those of other components (p < 0.001), whereas these parameters in CEL were recorded to be significantly lower (p < 0.001). Significant differences were observed in the degradation dynamics of CP: the rapid degradation phase was dominated by LS, while the slow degradation phase of PS was higher than that of other groups. In NDF, the rapidly degradable fraction of PR and LS was significantly higher than that of other groups (p < 0.001), with their ED recorded as 59.99% and 50.58%, respectively. In contrast, LBBL and CEL were identified to be significantly lower compared to the remaining groups (p < 0.001).

3.3. In Vitro Fermentation

3.3.1. In Vitro Gas Production

Figure 2 illustrates that all six raw materials show a gradual increase in gas production over time across substitution levels ranging from 0% to 100%. Among them, the GM and LBBL substitution groups exhibited significantly higher gas production at high substitution levels, suggesting a greater abundance of fermentable components, faster fermentation rates, and better potential degradability. Although PS and LS initially showed slower gas production, they accumulated considerable gas volumes by the 72 h mark. CEL demonstrated excellent continuous fermentation characteristics.

Based on the preliminary analysis of cumulative gas production, the PR and LS substitution groups maintained good gas production potential at a 100% substitution level, suggesting their suitability for high-level replacement. The GM substitution group achieved the highest gas production at a 50% substitution level, indicating that a moderate replacement level is more favorable for its fermentation. The PS, LBBL, and CEL substitution groups all performed best at a 75% substitution level.

3.3.2. In Vitro Fermentation Parameters

Figure 3 results indicate that the pH of all treatment groups remained within the suitable rumen range of 6.2 to 7.0. The LBBL–CEL substitution groups showed a slight increase in pH at high substitution levels but still maintained within the normal range. The microbial protein (MCP) concentration ranged from 16.57 to 33.31 mg/dL, with the LBBL–CEL substitution groups being slightly higher at high substitution levels, indicating an ideal supply of energy and nitrogen. Meanwhile, ammonia nitrogen (NH₃-N) significantly increased with the rise in substitution levels, reaching its peak at the 100% level in all substitution groups except for the PR and GM substitution groups.

The concentration of volatile fatty acids (VFAs) during in vitro fermentation displays distinct trends based on varying replacement ratios. In the PR group (

Table 8. Volatile Fatty Acid Compositions in In vitro Fermentation with Different Substrate Concentrations (mmol/L, n = 3).

Items		Group (%)					SEM	p-Value
		0	25	50	75	100
PR	HAc	16.34 e	21.53 d	22.49 c	24.40 b	30.82 a	1.21	<0.001
	PA	11.60 d	11.16 e	11.99 c	12.44 b	13.83 a	0.23	<0.001
	BA	9.10 b	8.74 d	8.02 e	8.99 c	9.76 a	0.15	<0.001
	A/P ratio, %	1.41 e	1.93 c	1.88 d	1.96 b	2.23 a	2.83	<0.001
	TVFAs	37.04 e	41.43 d	42.49 c	45.83 b	54.41 a	1.45	<0.001
GM	HAc	16.94 e	30.60 a	28.41 b	24.21 c	23.59 d	1.25	<0.001
	PA	12.60 ab	13.63 a	11.66 bc	10.84 cd	10.21 e	0.36	0.001
	BA	9.17 b	10.26 a	9.16 b	8.99 b	7.98 c	0.20	<0.001
	A/P ratio, %	1.34 c	2.26 ab	2.44 a	2.23 b	2.31 ab	2.80	<0.001
	TVFAs	38.71 e	54.48 a	49.22 b	44.05 c	41.77 d	1.51	<0.001
PS	HAc	16.94 d	34.99 c	43.97 a	38.04 b	38.11 b	2.46	<0.001
	PA	12.60 e	14.89 d	15.44 c	16.54 a	16.25 b	0.37	<0.001
	BA	9.17 a	8.01 e	8.10 d	8.94 b	8.80 c	0.12	<0.001
	A/P ratio, %	1.34 c	2.35 b	2.85 a	2.30 b	2.35 b	2.79	<0.001
	TVFAs	38.71 d	57.90 c	67.51 a	63.52 b	63.15 b	2.73	<0.001
LBBL	HAc	16.94 e	30.91 c	26.65 d	36.29 a	31.78 b	1.75	<0.001
	PA	12.60 e	14.32 c	13.05 d	16.96 a	15.71 b	0.44	<0.001
	BA	9.17 e	9.94 c	9.63 d	11.65 a	10.41 b	0.23	<0.001
	A/P ratio, %	1.34 e	2.16 a	2.04 c	2.14 b	2.02 d	2.82	<0.001
	TVFAs	38.71 e	55.17 c	49.34 d	64.90 a	57.91 b	2.35	<0.001
LS	HAc	16.94 e	22.52 d	25.28 c	30.49 b	35.13 a	1.68	<0.001
	PA	12.60 e	13.84 d	13.98 c	15.72 b	17.76 a	0.48	<0.001
	BA	9.17 d	6.40 e	9.59 c	9.92 b	11.13 a	0.42	<0.001
	A/P ratio, %	1.34 e	1.63 d	1.81 c	1.94 b	1.98 a	2.84	<0.001
	TVFAs	38.71 e	42.76 d	48.85 c	56.13 b	64.02 a	2.44	<0.001
CEL	HAc	16.94 c	33.59 a	30.72 b	31.22 b	30.60 b	1.59	<0.001
	PA	12.60 d	18.95 a	16.73 b	17.12 b	15.43 c	0.57	<0.001
	BA	9.17 d	11.87 a	11.24 b	11.28 b	10.93 c	0.25	<0.001
	A/P ratio, %	1.34 d	1.77 c	1.84 b	1.82 b	1.98 a	2.84	<0.001
	TVFAs	38.71 d	64.41 a	58.69 bc	59.62 b	56.95 c	2.37	<0.001

Different superscript letters indicate significant differences among treatments (p ≤ 0.05). PR: Pepper residue; GM: Grape marc; PS: Pepper straw; LBBL: Lycium barbarum branches and leaves; LS: licorice straw; CEL: Cyperus esculentus leaves. HAc: acetic acid; PA: propionic acid; BA: butyric acid; A/P ratio: acetate to propionate ratio; TVFAs: total volatile fatty acids.

), the concentration of acetic acid, propionic acid, butyric acid, the acetic-to-propionic acid ratio, and total acids exhibited a consistent increase with the increasing substitution ratio, with the 100% substitution group significantly higher than other groups (p < 0.001). The VFA contents in the GM group (Table 8) initially increased, followed by a decrease, with the 25% substitution group significantly higher than those in other groups (p < 0.001). In the PS group (Table 8), acetic acid and total acids reached their highest levels at a 50% substitution ratio (p < 0.001). The total acids in the LBBL (Table 8) and LS groups (Table 8) increased with the increased substitution ratio, with peaks observed at 100% and 75% substitution, respectively. Conversely, the CEL group (Table 8) displayed a gradual decrease in all VFAs indices as the substitution ratio increased, with the 25% substitution group showing significantly higher values than the other groups (p < 0.001).

4. Discussion

4.1. The Nutritional Components of Six Unconventional Feedstuffs

4.1.1. Nutritional Characteristics

PR exhibited the highest CP content, while CEL contained the lowest, indicating significant variance in protein supply capacity. The content differences in NDF and ADF in the six materials exceed 20%, it indicates that the protein levels in these different alternative feeds vary significantly and possess distinct fiber characteristics. Compared to traditional feeds such as alfalfa hay, PR and LBBL exhibit comparable crude protein content but relatively lower neutral detergent fiber and acid detergent fiber content, indicating their better potential degradability. The relatively low ADF in PS and moderate NDF suggest this material contains digestible fiber fractions, which may enhance rumen fermentation [23]. CEL, with its high ADF and NDF contents, is likely to be more resistant to microbial degradation, which was corroborated by its low rumen disappearance rate and fermentation efficiency.

The highest VC and VB1 contents were found in PR, and its antioxidant properties may indirectly promote fiber degradation through the protection of rumen microbial membranes’ integrity, as demonstrated in recent studies on agricultural by-products [24]. Higher levels of VC and VE were found in PS and LBBL, indicating potential for alleviating oxidative stress and improving milk quality, as supported by research on vitamin supplementation in dairy cows [25]. VA is generally low across all raw materials and feedstuffs, but the content of PS is relatively high, making it a potential auxiliary source of VA. Among the B vitamins, the high VB1 content in PR may enhance carbohydrate metabolic efficiency, while VB2 levels in GM and CEL may support protein and fat metabolism [26]. It should be noted that B vitamins are subject to substantial degradation and microbial synthesis in the rumen, which complicates the direct translation of their dietary content into post-ruminal availability. In conclusion, integrating feeds with high vitamin content can synergistically enhance the antioxidant and metabolic support functions of the diet, while reducing reliance on synthetic additives.

4.1.2. Amino Acid Profile and Protein Quality

Amino acids are the fundamental units of proteins, with lysine (Lys) and methionine (Met) considered the most limiting essential amino acids in ruminants. Studies have shown that supplementation with rumen-protected lysine (RP-Lys) and methionine (RP-Met) significantly enhances milk protein yield and milk fat content in dairy cows. Particularly under low-protein feed conditions, it can maintain or even increase milk production and milk composition while improving nitrogen utilization efficiency [27]. In addition, histidine (His) supplementation can also significantly improve the dry matter intake (DMI), milk yield, and true protein production in dairy ruminants [28]. In sheep, balanced amino acid supplementation helps optimize the colonic microbiota, improves lamb growth performance, and increases daily weight gain as well as final slaughter weight [29].

The highest levels of total essential amino acids and non-essential amino acids are detected in LS, with particularly elevated concentrations of leucine and valine, which are recognized as critical for muscle synthesis and nitrogen metabolism in ruminants. A significantly higher lysine content is observed in PS compared to other groups. As a limiting amino acid, it can enhance the efficiency of microbial protein synthesis. However, the lowest total amino acid content is recorded in CEL, which is associated with its low CP and high fiber levels. This nutritional profile suggests that CEL should be strategically combined with complementary protein sources in formulations to address its limitations, as emphasized in contemporary ruminant nutrition frameworks. The variability in amino acid content may be attributed to differences in plant tissue types and degrees of protein cross-linking, which affect digestibility and microbial accessibility in the rumen [30].

4.1.3. Polyunsaturated Fatty Acid Composition

Fatty acids play a crucial role in the energy metabolism of ruminants. Studies have found that supplementation with fats rich in palmitic acid (C16:0) can increase milk fat yield in dairy cows, while supplementation with stearic acid (C18:0) has a lesser effect on milk fat synthesis. Additionally, polyunsaturated fatty acids (PUFA), such as linoleic acid (C18:2) and α-linolenic acid (C18:3), are considered beneficial for improving meat quality and enhancing immune function [31].

GM has the highest unsaturated fat content among the six raw materials, especially linoleic acid (C18:2n6c) and oleic acid (C18:1n9c). These are healthy fats that can increase the good fat content in milk and meat. LS had the highest alpha-linolenic acid (C18:3n3) content, indicating potential benefits for enhancing n-3 PUFA levels in milk and meat. This aligns with studies suggesting that dietary inclusion of n-3-rich forages can improve ruminant product quality and reduce inflammatory responses [32]. CEL showed uniformly low PUFA content, supporting its classification as a low-energy, high-fiber material. The fatty acid differences across feeds could influence the biohydrogenation profile and methane emission in vivo, necessitating further research. It is important to note that a substantial proportion of these dietary PUFAs will undergo rumen biohydrogenation, thereby modifying their profile and potentially reducing their direct availability for absorption in the small intestine.

4.1.4. Functional Bioactives and Potential Effects

We detected low capsaicin content in PR but relatively high levels of capsaicin in PS, which is consistent with reported patterns of capsaicinoid accumulation in pepper stems. Studies have shown that adding capsaicin to dairy cow diets can improve metabolic status, enhance immune function, and increase milk yield. Furthermore, Capsaicin has been shown to modulate rumen microbial composition, potentially reducing protozoa and methane-producing archaea [33]. The proanthocyanidins detected in GM are a class of natural polyphenolic compounds with antioxidant and antibacterial effects. Studies have found that proanthocyanidins can improve rumen fermentation patterns in ruminants, reduce protein degradation, and enhance nitrogen utilization efficiency, thereby improving production performance [34]. Additionally, proanthocyanidins have been found to enhance immune function and improve animal health status. The Lycium barbarum polysaccharides abundant in LBBL possess multiple bioactive functions. Studies have shown that Lycium barbarum polysaccharides can improve the growth performance of ruminants, enhance antioxidant and immune capacities, and modulate the rumen microbial community [35]. Both liquiritin in LS and gallocatechin in CEL are bioactive ingredients in traditional Chinese herbal medicine, possessing certain anti-inflammatory effects. For instance, adding licorice extract to the diet of Karakul sheep enhances serum immunoglobulin levels and antioxidant capacity, although its impact on rumen fermentation parameters remains limited [36]. Epigallocatechin gallate (EGCG) possesses anti-inflammatory and antioxidant properties, which can protect bovine rumen epithelial cells from lipopolysaccharide-induced inflammatory damage by activating the autophagy pathway, thereby improving rumen health [37].

4.2. Rumen Degradation and In Vitro Fermentation

Rumen degradability has been observed to be modulated by the combined effects of substrate chemical composition, cell-wall microstructure, lignin bonding configuration, and bioactive ingredients. In this study, significant variation in DM, NDF, and ADF among six agricultural by-products was linked to marked differences in their degradation kinetics.

PR exhibited the highest DM effective degradability and rapidly degradable fraction, despite having a moderate NDF content. This observation aligns with findings that fiber degradability is more dependent on cell-wall structure and lignin association than on NDF content alone. Jung and Grabber further emphasized that cross-linked cell-wall structures act as physical barriers to microbial penetration [38]. In this study, PR was likely characterized by a lower ADF content, suggesting a less lignified cell wall structure, along with a more porous architecture and higher levels of non-structural carbohydrates. These factors collectively contributed to its enhanced degradability [39].

Conversely, CEL, with the highest NDF and ADF contents, showed the lowest degradability values for DM and fiber, as well as a minimal rapidly degradable fraction. This is consistent with previous reports on the poor ruminal degradability of by-products with high fiber content such as corn cobs [40] but contrasts with studies reporting efficient fiber digestion of Cyperus species in pigs or enzymatic systems. This discrepancy may result from the higher sensitivity of rumen microbiota to flavonoids, which have been shown to suppress fibrolytic activity [41].

Additionally, PS exhibited a high proportion of slow degradation but only moderate effective degradability, suggesting the presence of potentially degradable fiber that remained unreleased during short-term fermentation. This pattern is similar to the “latent degradability” phenomenon reported for alfalfa stems. In summary, compared to conventional forages such as rice straw and corn straw, by-products like PR and LS demonstrated significantly higher degradation potential. However, their effective utilization requires compatibility with ruminal microbiota and structural adaptability. Recent research has increasingly emphasized the role of structural–functional relationships over simple compositional analysis [38], and the present study reinforces this perspective by elucidating the structural basis of substrate degradability differences.

PR and LS were observed to maintain high gas output and VFAs production at 100% replacement levels, indicating that their degradation products were readily utilized by rumen microbiota. This is consistent with the conclusion reported by An et al. that “the abundant polyphenols in chili by-products have a positive impact on rumen fermentation [42].

Conversely, GM exhibited the classic inverted “U”-shaped response: maximal gas production was achieved at 50% inclusion, whereas VFA concentration peaked at 25%, declining beyond this level. This phenomenon corresponds with the dose-dependent biphasic effects of proanthocyanidins described by Singh et al., where moderate doses enhance nitrogen utilization and high doses suppress fibrolytic enzyme expression [34]. Suescun-Ospina et al. further reported that tannins in grape pomace may bind proteins and interfere with metabolic pathways, ultimately reducing nitrogen efficiency [43].

LBBL demonstrated maximal MCP synthesis at 75% substitution, while maintaining stable VFAs and NH₃-N levels, suggesting that its abundant polysaccharides facilitated carbon–nitrogen fermentation synergy—a finding consistent with reports that medicinal polysaccharides enhance rumen MCP production [7].

Except for PR, nearly all substrates demonstrated non-linear responses between inclusion level and fermentation efficiency, with distinct optimal thresholds observed. For example, CEL reached a peak total VFA concentration of 64.41 mmol/L at 25% substitution but exhibited declining VFAs levels, rising pH, and increased NH₃-N at higher inclusion rates. This suggests that, at elevated concentrations, the combined effects of fiber structure and plant secondary metabolites may disrupt rumen microbial balance. Such observations are consistent with the findings of Patra & Saxena, who indicated that plant secondary metabolites can exert synergistic or antagonistic effects depending on their concentrations and should thus be carefully managed in feed application [44].

5. Conclusions

Based on a comprehensive evaluation of ruminal degradability, VFA and MCP production, and ammonia-N accumulation, optimal substitution levels were identified for each substrate. The optimal inclusion levels within the diet, on a dry matter basis, were determined as follows: the inclusion of PR was deemed most favorable at 150–200 g/kg; GM showed optimal outcomes at 50–100 g/kg; PS performed best at an inclusion level of 100 g/kg; the optimal level for LBBL was determined to be 150 g/kg; LS was best utilized within a range of 150–200 g/kg; and the inclusion of CEL should not exceed 50 g/kg. These tailored recommendations enable substrate-specific feed formulation aligned with structural attributes and functional compound profiles, thereby optimizing rumen fermentation efficiency and nitrogen utilization.