Effects of Different Molasses Levels and Slow-Release Urea Combinations on Growth Performance, Serum Biochemistry, Rumen Fermentation, and Microflora of Holstein Fattening Bulls

Abstract

The aim of this study was to investigate the effects of different combinations of molasses levels and slow-release urea on Holstein fattening bulls. Sixty Holstein fattening bulls of a similar age, weight, and health status were randomly divided into four groups of fifteen Holstein fattening bulls each. All of the treatments were as follows: (1) basic diet group (CON); (2) 2% molasses slow-release urea group (LMU); (3) 4% molasses slow-release urea group (MMU); and (4) 6% molasses slow-release urea group (HMU). The results of the study showed that the MMU had an outstanding performance, with a 13.3% increase in average daily weight gain compared with the control group, a significant decrease in feed conversion ratio (p < 0.05), and a significant increase in apparent digestibility of crude protein (p < 0.05). In terms of serum biochemical indices, blood ammonia and alanine aminotransferase (ALT) concentrations were significantly higher in the MMU than in the CON (p < 0.05). The rumen pH of all treatment groups was lower than that of the CON (p < 0.05), whereas the concentrations of microbial crude protein (MCP), as well as acetic acid, propionic acid, and total volatile fatty acids (TVFA) were significantly higher in both the MMU and HMU (p < 0.05). The dominant phyla in each group were Bacteroidetes, Firmicutes, and Patescibacteria, and the relative abundance of Bacteroidetes in the MMU increased by 5.47% compared with that in the CON. In the MMU, Prevotellaceae, Lachnospiraceae, and Ruminococcaceae increased by 9.03%, 0.67%, and 3.43%, respectively, compared with the CON. The economic benefit analysis showed that the daily feeding cost of fattened cattle in the MMU was reduced by RMB 1.62 yuan, and the daily farming benefit of each cow was increased by RMB 7.19 yuan. In conclusion, the MMU was effective in improving the growth performance of fattening cows, optimizing rumen fermentation, reducing cost, and increasing profit, which is a nutritional strategy with great application value.

1. Introduction

Protein in feed provides essential amino acids and nitrogen sources for rumen microorganisms in ruminants; however, dietary proteins are usually derived from higher-cost feeds, such as soybean meal. Non-protein nitrogen has been widely used in ruminant diets to replace some of the feed proteins, due to its low cost [1]. Nitrogen in feed is often not fully utilized by rumen microorganisms due to asynchronous supply and energy release, leading to increased urinary nitrogen emissions that are converted to ammonia or nitrous oxide, increasing greenhouse gas emissions. Urea-dissolving bacteria are the most important organisms in the rumen that produce the enzyme urease, which breaks down urea into NH₃ for microbial protein synthesis [2]. The hydrolysis of urea to produce NH₃ is rapid and exceeds its utilization by rumen microorganisms for the production of microbial proteins, causing the release of energy nitrogen that is not synchronized, and leading to nitrogen waste and environmental pollution [3,4]. When the NH₃-N concentration exceeded 30 mg/L, the ammonia utilization efficiency was significantly reduced, and the excess ammonia was absorbed into the bloodstream and converted into blood ammonia, which may lead to hyperammonemia and metabolic burden on the liver. Further studies found that high concentrations of NH₃ also indirectly reduced feed utilization by affecting rumen pH balance and inhibiting the activity of rumen fibre-degrading bacteria [5]. Therefore, in order to optimize the rumen environment and animal health, it is necessary to regulate the nitrogen supply through precise feed formulation, to keep the NH₃-N concentration within the appropriate range. Urea is hydrolyzed and releases excess ammonia (NH₃) into the gastrointestinal tract, where it is absorbed and causes hyperammonaemia [6]. Therefore, balancing hydrolysis rates with NH₃ assimilation rates is crucial for the efficient ruminant utilization of any nitrogen. Different methods have been employed to slow down urea hydrolysis rates, such as increasing the energy release rate and slowing down the release of urea [7].

Previous studies found that with an increase in dietary total sugar content, the rumen ammonia nitrogen concentration decreased linearly, and the rumen urea loss rate decreased significantly [8,9]. Compared with other carbon sources (such as starch or cellulose), molasses can provide a large amount of readily available energy in a short period of time, which can effectively improve the nitrogen utilization rate and protein synthesis rate of microorganisms [10]. The improvement of fermentation status increased the demand of NH₃-N for rumen bacteria [11]. To fully utilize urea and maintain rumen microbial efficiency, starch- and molasses-based supplements have been used to achieve energy nitrogen balance in goats [12,13].

Many studies have confirmed that urea utilization in ruminants can be improved by adjusting feeding strategies. However, the use of molasses as a source of energy supplementation to improve urea nitrogen utilization in Holstein fattening cattle has not been studied. Therefore, the aim of this experiment was to investigate the effects of adding different levels of molasses and slow-release urea on the performance, serum biochemical indices, and rumen microbiota of Holstein fattening cows. Although the results of the study contributed to the understanding of rumen function, there is still a need for an in-depth study of their effects on the meat quality of breeding bulls. However, differences in the source of molasses and the preparation process may affect the generalizability of the results, and due to the short experimental period, the long-term and cumulative effects of the two products need to be further investigated. Nevertheless, we believe that appropriate doses of molasses and slow-release urea can regulate the structure of rumen microbial communities and improve feed utilization efficiency, thereby promoting growth.

2. Materials and Methods

2.1. Animals, Diets, and Experimental Design

This experiment was approved by the Committee of the College of Animal Science and Technology, Anhui Agricultural University (No. SYDW-P20190600601). The trial period was from 28 January 2024 to 26 February 2024. Sixty Holstein fattening bulls of a similar age (13.5 ± 2 months), body weight (450 ± 30 kg), and health status were randomly selected and divided into four groups, with fifteen Holstein fattening bulls in each group: (1) the control group (CON): the group was fed TMR without molasses and slow-release urea; (2) low molasses slow-release urea group (LMU): use of 2% (of TMR dry matter) molasses (41% total sugar content, Heli Co., Ltd., Weifang, China) and 3% (of TMR dry matter) gelatinized starch urea (70% urea, Huarui, Co., Ltd., Hefei, China) replaced isocaloric and isonitrogenous portions in TMR; (3) medium molasses slow-release urea group (MMU): 4% (of TMR dry matter) molasses and 3% (of TMR dry matter) gelatinized starch urea were used to replace isocaloric and isonitrogenous portions in TMR; and (4) high molasses slow-release urea group (HMU): 6% (of TMR dry matter) molasses and 3% (of TMR dry matter) gelatinized starch urea were used to replace isocaloric and isonitrogenous portions in TMR. The pre-test lasted 7 days and the entire trial 30 days. Holstein fattening bulls were given the diet ad libitum, with a target refusal rate of 5%. The experimental groups were fed the TMR diet twice a day at 9:00 and 16:00, and the cattle were fed freely and drank freely. Every afternoon, the manure accumulated in the cattle barn was cleaned and the cattle barn was sprayed for disinfection. All diets were formulated in accordance with the NRC (2001) and the Chinese Ministry of Agriculture’s Dairy Cattle Feeding Standard (NY/T 34-2004).

Table 1. TMR composition and nutritional level (% of DM, unless otherwise stated).

Items	Content (%)
	CON	LMU	MMU	HMU
Corn silage	55.60	55.60	55.60	55.60
Straw	8.30	8.30	8.30	8.30
Corn	21.66	27.30	27.51	27.94
Wheat	5.41	3.83	2.71	1.30
Soybean meal	6.86	1.01	1.19	1.44
Molasses	-	0.71	1.44	2.17
Gelatinized starch urea	-	1.08	1.08	1.08
NaHCO3	0.36	0.36	0.36	0.36
Premix 1	1.81	1.81	1.81	1.81
Total	100	100	100	100
Nutrients
ME (MJ/kg) 2	5.56	5.56	5.56	5.56
DM	48.17	48.19	48.13	48.16
NDF	36.48	36.29	36.15	36.91
ADF	17.87	17.63	17.70	17.71
CP	15.43	15.41	15.40	15.45
EE	2.19	2.13	2.17	2.23

Abbreviations: TMR = total mixed ration; DM = Dry matter; NDF = Neutral detergent fibre; ADF = Acid detergent fibre; CP = Crude protein; EE = Ether extract. ¹ Each kg of premix contains Lysine 8000 mg, VA 60,000 IU, VD 60,000 IU, Fe 500 mg, Cu 200 mg, Zn 500 mg, Mn 400 mg, I 5 mg, and Se 2 mg. ² Nutrient composition of diets is measured except for metabolizable energy, which is calculated. All four groups of TMR diets followed the isoenergetic and isonitrogenous principles.

shows the TMR composition and nutrient levels.

2.2. Feed Analysis

During the test period, TMR samples were collected on 2 consecutive days per week using the 5-point sampling method, and the collected samples were mixed, dried at 65 °C for 48 h, crushed and mixed homogeneously, and sealed in self-sealing bags to determine the nutrient composition of the feeds. Before the end of the experiment, 10 Holstein fattening bulls were randomly selected from each group and fecal samples were collected through the rectum on Days 30 and 31. The fecal samples collected for 2 days were divided into two parts. 10 mL 10% HCl was added into every 100 g of fresh feces, part of the sample was treated with nitrogen fixation, and the other part of the sample was mixed evenly and frozen at −20 °C for subsequent determination of apparent digestibility. The standard procedure of the Association of Official Analytical Chemists [14] was used to analyze the DM (Dry matter), OM (Organic matter), Crude protein (CP) and ash content of the feed. Neutral detergent fibre (NDF) and Acid detergent fibre (ADF) were measured according to Van Soest et al. [15].

2.3. Blood Sampling and Analysis

Ten Holstein fattening cattle in each group were selected and blood was collected through the tail root vein before morning feeding on Day 30 of the experimental period, and placed in 10 mL non-anticoagulant vessels (Kangwei Shi Medical Technology, Co., Ltd., Shijiazhung, China). Blood was allowed to stand until the supernatant was separated, and was then centrifuged (3500 rpm/min, 15 min, Shanghai Lu Xiangyi Centrifuge Instrument, Co., Ltd., Shanghai, China). After centrifugation, the serum and plasma were separated into 1.5 mL centrifuge tubes (Beijing Labgic Technology Co., Ltd., Beijing, China) and stored at −20 °C. Measurements included total protein (TP), albumin (ALB), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), lactic dehydrogenase (LDH), urea nitrogen, blood urea nitrogen (BUN), ammonia (AN), urea acid (UA), and creatinine (CREA). Blood samples were tested by Beijing Huaying Biotechnology.

2.4. Rumen Fluid Sampling and Analysis

On Day 30 of the experimental period, 2 h after the bulls were fed, 6 cows in each group were selected to collect rumen fluid. Rumen fluid was collected using a rumen fluid sampler (Kelibo, Co., Ltd., Wuhan, China), and a sampling catheter was inserted into the rumen through the mouth and esophagus; 100 mL of rumen fluid collected at the beginning of the collection was discarded, and the sample continued to be collected for approximately 200 mL of rumen fluid. The pH of the rumen fluid samples was measured immediately after collection using a portable pH metre (Yidian Scientific Instrument Co., Ltd., Shanghai, China). Each rumen fluid sample was filtered through four layers of gauze, and one portion of the sample was stored in a refrigerator at −20 °C for analysis of NH₃-N, microbial protein (MCP), and volatile fatty acid (VFA), respectively. The other portion was quickly packed into cryopreservation tubes and preserved in liquid nitrogen tanks. It was subsequently transferred to −80° C for freezing and preservation and used for the determination of rumen bacteria 16S rRNA. The content of MCP was determined by the Kaumas Brilliant Blue method, the absorbance of the filtered fermentation liquid was determined by spectrophotometer, and the content of MCP was calculated by a regression formula based on the binding property of Kaumas Brilliant Blue G-250 dye and protein [16]. Ammonia nitrogen was measured using the phenol–hypochlorite method [17].

A GC-2010 gas chromatograph (Shimadzu, Japan) was used to determine the volatile fatty acids in the rumen fluid. Rumen fluid treatment was conducted as follows: The rumen fluid was centrifuged at 12,000 rpm for 10 min, and then 1 mL of the supernatant was added to 0.2 mL of crotonic acid–metaphosphoric acid mixed solution, mixed well, filtered again through the needle-type filter membrane, and then automatically injected into the sample for analysis. Determination conditions: the column was a wonda cap wax capillary column (30.0 mm in length, 0.25 μm in film thickness, and 0.25 mm in inner diameter). The temperature of the gasification chamber was 200 °C, and the temperature of the FID detector was 220 °C. The column temperature was determined by the programmed temperature increase method, with an initial temperature of 80 °C, an end temperature of 170 °C, and an increase rate of 15 °C/min, and crotonic acid was used as the internal standard.

2.5. Growth Performance Measurement

Weighing was performed once at the beginning of the experiment period and once at the end of the experiment period, each time before the morning feeding. During the experiment period, the feed intake of experimental cattle was recorded for 2 consecutive days a week, and the average daily gain and feed-to-gain ratio were calculated.

Average daily gain (ADG) = (final weight − initial weight)/test days

Daily feed intake (DMI) = amount of feed − residual feed amount

Feed-to-weight ratio (F/G) = total feed consumption/total weight gain

2.6. Analysis of Bacterial Community in Rumen

Genomic DNA was extracted from the rumen fluid using a FastDNA Spin Kit for Soil (Meiji Bio Ltd., Shanghai, China) after thawing 1 mL of rumen fluid at 4 °C. The concentration and purity of the genomic DNA were determined using an ultra-micro spectrophotometer (NanoDrop2000, Thermo Fisher Scientific, Waltham, MA, USA). Fragments of the V3–V4 region of the bacterial universal 16S rRNA gene were selected for this purpose, and the bacterial upstream primer 338F (5′-ACTCCTACGGGGAGGCAGCAG-3′) and downstream primer 806R (5′-GGACTACHVGGGTWTCTAAT3′) for PCR amplification. The PCR products were identified, purified, and quantified by 2% agarose gel electrophoresis, and purification was performed using the DNA Gel Recovery and Purification Kit (PCR Clean-Up Kit, Passover, Shanghai, China). The PCR products were detected and quantified using a Qubit 4.0 (Thermo Fisher Scientific, USA). Libraries were constructed using the NEXTFLEX Rapid DNA-Seq Kit, and the resulting amplified fragments were sequenced on Illumina’s Miseq PE300 sequencing platform. The raw sequences were quality-controlled using fastp software (version 20.0) and spliced using FLASH software (version 1.2.7). Using UPARSE software (version 11), according to the 97% similarity of the sequence for Operational Taxonomic Units (OTU) clustering, the clustered sequences after chimera filtering were used to obtain the OTU representative sequences used for species classification. Taxonomic annotation of OTU species was performed using an RDP classifier (version 2.11) with a confidence threshold of 70%, and the community composition of each sample was counted at different species taxonomic levels (Phylum, Family, Genus). All data analyses were performed on the Meggie BioCloud platform (https://cloud.majorbio.com, accessed on 11 March 2024), as described below.

The diversity analysis mainly includes alpha diversity and beta diversity. Alpha diversity contains two factors of richness and evenness of species composition in the samples, which are usually evaluated by the ACE index, Chao1 index, Shannon index, and Simpson index. Beta diversity compares the microbial community composition of different samples. Here, the differences between samples of different treatments were mainly demonstrated by principal coordinate analysis (PCoA). The vegan package in R was used for the PCoA analysis of rumen microbial communities based on the Bray–Curtis distance algorithm, and visualized with the ggplot2 package. Similarity analysis was also performed with the vegan package (ANOSIM) to indicate community similarity. Analysis of differences between groups: Microbial component differences were analyzed primarily by linear discriminant analysis (LDA), effect size (LEfSE), and secondarily by LDA to estimate the magnitude of the effect of the abundance of each species on the differential effect.

2.7. Statistical Analysis

Statistical analyses were performed using SPSS (version 26, IBM, USA) to assess between-group differences in growth performance, serum parameters, rumen parameters, and the bacterial abundance of Holstein fattening cows using different feeding combinations. The normality of the distribution of the variables was assessed using the Shapiro–Wilk test, which showed that no variables significantly deviated from the normal distribution (p > 0.05) and the assumption of normality was satisfied. Subsequently, homogeneity of variance was tested by Levene’s Test and the results showed that the assumption of homogeneity of variance was satisfied (p > 0.05). On this basis, changes in the dependent variables under different feeding combinations were assessed using one-way analysis of variance (ANOVA), with the level of significance set at p < 0.05. If significant differences were observed, multiple comparisons were performed using Duncan’s post hoc test to identify differences between groups.

3. Results

3.1. Growth Performance

The growth performance results are shown in

Table 2. Effects of different molasses and slow-release urea on growth performance.

Items	CON	LMU	MMU	HMU	SEM	p-Value
Initial Weight (kg)	424.69	438.93	442.73	440.40	4.968	0.609
Final Weight (kg)	502.62	517.53	531.07	437.1	5.148	0.298
ADG (kg)	1.66 b	1.67 b	1.88 a	1.71 b	0.015	<0.001
DMI (kg/d)	11.46	11.70	11.73	11.60	0.048	0.206
DMI/ADG	6.92 a	7.04 a	6.25 b	6.81 a	0.055	<0.001

Abbreviations: ADG = average daily gain; DMI = dry matter intake. ^a and ^b indicate a significant difference at p < 0.05, while the same letter indicates no significant difference between groups.

. The ADG and FCR of the MMU were significantly higher than those of CON and other treatment groups (p < 0.05), and the DMI of the MMU was not significantly different from that of the CON and other treatment groups.

3.2. Nutrient Digestibility

Nutrient digestibility is shown in

Table 3. Effect of different molasses and slow-release urea combinations on digestibility coefficient.

Items	CON	LMU	MMU	HMU	SEM	p-Value
DM, %	86.60	86.59	86.74	86.81	0.105	0.865
OM, %	65.68 b	66.78 a	66.30 ab	66.63 a	0.139	0.020
CP, %	62.01 c	63.65 bc	66.15 a	64.81 ab	0.441	0.004
NDF, %	49.95	50.88	50.08	51.37	0.736	0.900
ADF, %	28.26	28.23	29.43	28.61	0.733	0.938
EE, %	72.71	72.85	72.86	72.64	0.096	0.832

Abbreviations: DM = dry matter; OM = organic matter; CP = crude protein; NDF = Neutral detergent fibre; ADF = acid detergent fibre; EE = ether extract. ^a–c Values with different letters in the same row are significantly different (p < 0.05).

. The apparent digestibility of OM and CP was higher than that of the CON in all treatment groups. The apparent digestibility of OM was significantly higher than that of the CON in both the LMU and HMU (p < 0.05). The apparent digestibility of CP was significantly higher than that of the CON and LMU in the MMU (p < 0.05), which was not different from that of the HMU. The differences in the apparent digestibility of DM and EE among the experimental groups were not significant.

3.3. Serum Biochemistry

The test results of the serum indexes are shown in

Table 4. Effect of different combinations of molasses and slow-release urea on serum biochemistry.

Items	CON	LMU	MMU	HMU	SEM	p-Value
TP (g/L)	66.86	66.17	65.30	66.23	0.655	0.880
ALB (g/L)	29.77	29.83	29.27	29.28	0.364	0.919
BUN (mg/dL)	4.13 b	4.03 b	4.30 b	5.39 a	0.184	0.025
AN (µmol/L)	27.01 a	26.43 ab	24.99 b	24.66 b	0.342	0.038
ALT (U/L)	30.52 a	28.45 ab	25.92 b	25.43 b	0.724	0.044
AST (U/L)	81.01 b	92.33 a	78.42 b	79.26 b	1.942	0.029
ALP (U/L)	175.45	189.93	158.77	178.41	7.341	0.517
CREA (µmol/)	77.55 b	84.09 a	73.30 ab	79.14 ab	1.381	0.039
LDH (U/L)	1029.25	1100.31	989.88	1004.20	18.419	0.139

Abbreviations: TP = total protein; ALB = albumin; BUN = blood urea nitrogen; AN = ammonia; ALT = alanine aminotransferase; AST = aspartate aminotransferase; ALP = alkaline phosphatase; CREA = creatinine; LDH = lactic dehydrogenase. ^a and ^b indicate a significant difference at p < 0.05, while the same letter indicates no significant difference between groups.

. The BUN concentration in the HMU was significantly higher than that in the other three groups (p < 0.05). The concentration of AN and ALT in both the MMU and HMU was significantly lower than that in the CON (p < 0.05). The concentration of AST and CREA in the LMU was significantly higher than that in the CON (p < 0.05), and there were no significant differences in TP, ALB, UA, and LDH between the test groups.

3.4. Rumen Fermentation

The results of the rumen fluid tests are shown in

Table 5. Effect of different molasses and slow-release urea combinations on rumen fermentation.

Items	CON	LMU	MMU	HMU	SEM	p-Value
pH	6.80 a	6.67 ab	6.66 b	6.50 c	0.031	0.002
MCP (mg/dL)	154.44 b	170.81 ab	176.73 a	183.95 a	3.700	0.022
NH3-N (mg/dL)	15.69	17.63	13.65	18.71	1.343	0.588
TVFA (mmol/L)	87.41 c	91.64 bc	97.43 ab	100.91 a	1.610	0.006
Acetate (mmol/L)	59.72 c	63.76 bc	67.03 ab	70.25 a	1.226	0.008
Propionate (mmol/L)	16.70 b	16.68 b	18.92 a	19.09 a	0.397	0.024
Butyrate (mmol/L)	10.74	10.98	11.23	11.33	0.880	0.067
Isobutyrate (mmol/L)	0.05	0.05	0.04	0.05	0.001	0.368
Valerate (mmol/L)	0.12	0.10	0.13	0.12	0.004	0.093
Isovalerate (mmol/L)	0.08	0.07	0.07	0.09	0.003	0.285
Acetate/Propionate	3.59	3.86	3.54	3.69	0.059	0.231

^a–c Values with different letters in the same row are significantly different (p < 0.05).

. Ruminal pH was significantly lower (p < 0.05) in all treatment groups than in the CON. MCP content was significantly higher (p < 0.05) in all treatment groups than in the CON, and acetic acid, propionic acid, and TVFA content were significantly higher (p < 0.05) in the MMU and HMU than in the CON. There were no significant differences in NH₃-N, butyric acid, isobutyric acid, valeric acid, isovaleric acid, or propionic acid acetate ratios among the test groups.

3.5. Rumen Microbiota Diversity Analysis

The estimated α diversity index of the rumen 16S rRNA gene library of Holstein fattening bulls obtained by sequencing analysis is shown in Figure 1. The results indicated that the Chao 1 index and Shannon index were significantly lower (p < 0.05), while the Simpson index was significantly higher (p < 0.05) in each treatment group compared to the control group. There were no significant differences between the different treatment groups for bacterial community coverage and the Ace index.

To compare the differences in bacterial colony structure between treatment groups, the differences between treatment groups were compared more intuitively by means of a box plot based on the principal coordinate analysis (PCoA) of the Bray–Curtis distance algorithm. As shown in Figure 2, the differences in community composition between the CON and MMU were large, while the community composition of the LMU and HMU was more similar to that of both the CON and MMU (between the two).

3.6. Analysis of the Relative Abundance of Rumen Bacteria

Based on 2218 OTUs, a total of 20 phyla, 43 orders, 106 orders, 195 families, and 402 genera of bacteria were identified in this experiment. The bar diagram shows the trend of species changes in the microbial communities of different subgroups, while the heatmap diagram shows the microbial communities of the dominant species in each group, which is complementary to the bar diagram.

At the phylum level, the dominant phyla of rumen bacteria in beef cattle were Bacteroidetes, Firmicutes, and Patellobacteria, with relative abundances of 51.52%, 38.91%, and 3.90%, respectively, in the CON, 54.98%, 39.21%, and 2.66%, respectively, in the LMU, 56.99%, 37.65%, and 1.64%, respectively, in the MMU, and 53.16%, 39.98%, and 2.47%, respectively, in the HMU (Figure 3a).

At the family level, Prevotellaceae (31.40%, 37.30%, 40.43%, 31.71%), Lachnospiraceae (8.86%, 9.78%, 9.53%, 12.03%), Ruminococcaceae (4.02%, 7.61%, 7.62%, 7.33%), Bacteroidales_RF16_group (5.22%, 5.85%, 6.83%, 8.41%), and Rikenellaceae (6.90%, 5.99%, 4.71%, 6.90%) were in the CON and LMU, 4.71% and 6.90%, respectively, with the MMU and HMU groups as dominant bacterial families (Figure 3c).

At the genus level, Prevotella (25.98%, 32.51%, 35.41%, 26.51%), the norank_f__Bacteroidales_RF16_group (5.19%, 5.87%, 6.84%, 8.40%), the Rikenellaceae_RC9_gut_group (6.65%, 5.87%, 4.62%, 6.76%), Ruminococcus (3.20%, 6.90%, 6.63%, 6.34%), Succiniclasticum (5.9%, 5.07%, 2.80%, 4.22%), Clostridium spp. UCG-014 (3.65%, 3.65%, 3.95%, 2.78%), and norank_f__F082 (2.94%, 2.49%, 1.74%, 3.38%) were the dominant genera of the CON, LMU, MMU, and HMU (Figure 3e).

3.7. LEfSe and Significant Difference Species Enrichment Analyses

Based on the community abundance data obtained, hypothesis testing was conducted by the Kruskal–Wallis rank sum test for species between microbial communities of different treatment groups, to assess the significance level of species abundance differences, and then to obtain significantly different species between groups. As can be seen in Figure 4a, at genus level, NK4A214_group, norank__f__p-251-o5, norank__f__norank__o__Absconditabacteriales_SR1, and Fibrobacter were significantly elevated in the CON compared to the other groups, while in the MMU, Colidextribacter and Selenomonas were significantly elevated. Species differences in species at different taxonomic levels are shown in the form of developmental dendrograms, which visually reflect the differential species at the level of different species hierarchies obtained between the different treatment groups. The LEfSe shown in Figure 4b,c mainly analyzed the treatment groups (CON, LMU, MMU, and HMU) and yielded a total of 38 significantly different microbiota, with an LDA threshold of 3. Of these, 32 species were significantly enriched and the CON 4 microorganisms were significantly enriched in the HMU. The following phyla were determined to be in the CON: Cyanobacteria (p = 0.002, LDA: 3.35), Fibrobacterota (p = 0.016, LDA: 3.28), Patescibacteria (p = 0.032, LDA: 4.08), and Verrucomicrobiota (p = 0.026, LDA: 3.57). At the genus level, the relative abundance of Fibrobacter (p = 0.016, LDA: 3.28) and UCG-010 (p = 0.049, LDA: 3.13) phyla was significantly higher, and the relative abundance of Colidextribacter (0.032, LDA: 3.85) and Selenomonas (p = 0.01, LDA: 3.28) phyla was significantly higher in the MMU (LDA: 3.28). Genera were significantly higher in relative abundance and the Acetobacter genera were significantly higher in the HMU (p = 0.001, LDA: 3.14).

3.8. Economic Benefit Analysis

Based on the calculation of raw material prices for pasture in January 2024, the price of finished corn silage is 600 yuan/ton, rice straw is 500 yuan/ton, corn is 2300 yuan/ton, wheat bran is 1750 yuan/ton, and soybean meal is 3600 yuan/ton. Additionally, the on-site price of molasses is 2000 yuan/ton, slow-release urea is 2400 yuan/ton, baking soda is 2300 yuan/ton, and beef cattle premix is 3200 yuan/ton. The analysis of comprehensive economic benefits, considering the varying daily weight gains of each group, indicates that incorporating 4% molasses and 3% slow-release urea into the diets of fattening cattle—replacing a portion of the concentrates—yields the highest economic advantage. This modified diet results in a feed cost that is lower than that of the CON, saving 1.62 yuan per day per head. Additionally, the breeding income increases by 7.19 yuan per day per head compared to the CON, as illustrated in

Table 6. Economic benefit analysis table of each experimental group.

Items	CON	LMU	MMU	HMU
Weight gain (kg/day/head)	1.66	1.67	1.88	1.71
Unit price of live cattle (yuan/kg)	25.12	25.12	25.12	25.12
Weight gain income (yuan/day/head)	41.65	42.00	47.23	42.83
Feed cost (yuan/day/head)	33.31	31.53	31.69	31.92
Feed cost per unit weight gain (yuan/day/head)	20.09	18.86	16.86	18.72
Breeding income (Yuan/day/head)	8.34	10.47	15.53	10.91

.

4. Discussion

In this experiment, there was no significant difference in dry matter intake among all groups, but a significant difference in the digestibility of organic matter and crude protein. Chibisa et al. found that adding pure lactose to a barley or corn-based diet had no significant effect on the DMI of dairy cows [18], and Seram et al. found the same result [11]. Silvia Grossi et al. found that slow-release urea, replacing part of the soybean meal, increased the digestibility of crude protein and fibre with no significant effect on DMI [19]. A meta-analysis to assess the effect of slow-release urea in dairy production showed no significant effect of DMI, and a significant increase in the digestibility of both DM and CP [20], which is consistent with the results of the present trial. The lack of difference in DMI usually implies that feeding behaviour is stable across feed treatments, whereas the increase in digestibility suggests that ruminants are able to better utilize the nutrients that they would have consumed at a similar level of intake, with increased production performance, improved health, and a reduced environmental impact. Adejoro et al. studied the effects of adding urea and nitrate to fattening lambs, and found that the dry matter intake of lambs was not affected by nitrogen source [21]. This is consistent with the results of this experiment. Broderick et al. also found that when corn silage-based Holstein dairy cows were fed sucrose instead of corn starch, the rapidly degrading carbon source sucrose increased the milk yield and milk protein content. This indicates that the matching degree of dietary sucrose and nitrogen sources is higher than that of corn starch [22]. Syamsi et al. found that a treatment group with good nitrogen release synchronization had a significant impact on the digestibility of DM and OM in goats [23]. With the improvement of the digestibility of OM and CP, more nutrients were digested, absorbed, and utilized, thus increasing the average daily gain of each treatment group. The simultaneous degradation rate of carbon sources and nitrogen sources in the MMU helps animals to improve the efficiency of protein utilization, use energy and nitrogen sources in food more effectively, and reduce nitrogen waste.

ALT and AST decreased significantly in the MUN group, which may be because the nitrogen source released in the MUN group can be better utilized to synthesize MCP, which also echoes the significant increase in MCP content in the MUN group. With the decrease in blood ammonia, the ammonia entering the liver is also relatively reduced, the liver burden is reduced, and the metabolic waste of ALT and AST produced by the liver are also reduced. Interestingly, this is similar to the results of Dolatkhah et al.’s study [24]. Research has shown that feeding cows with high concentrations may lead to increased BUN content [25]. This is consistent with the results of this experiment. Compared with other treatment groups, the BUN content in the HMU was significantly increased, which may be due to the disturbance of the rumen microbial metabolism caused by the increase in molasses content.

Syamsi et al. found that the rumen pH of goats in the treatment group with good nitrogen release synchronization was significantly reduced [23]. In vitro studies by LEE et al. found that pH decreased with the increase in water-soluble carbohydrates (WSC) after adding a mixture of WSC (inulin and sucrose; 80:20) to ryegrass [26]. Fluctuations in rumen pH directly affect fermentation efficiency and animal health, while increases in volatile acids are accompanied by decreases in pH [27]. This is consistent with the results of the present experiment, where pH was lower in all experimental groups than in the control group, corresponding to an increase in rumen fermentation VFA production. NH₃-N is the most important source of MCP synthesis, and the MCP synthesis rate is also affected by fermentable CHO. When energy is limited or NH₃-N is insufficient, MCP cannot be effectively synthesized, resulting in the accumulation of ammonia in the rumen and an increase in the level of ammonia [28]. The MCP content of all treatment groups was higher than that of the CON, and the MCP content of the MMU and HMU was significantly higher than that of the CON. The synchronization of energy and nitrogen release significantly affects microbial protein synthesis [29]. In the MMU, the synchronization of energy nitrogen release is better, more ammonia is effectively used to produce MCP, and the ammonia nitrogen content is lower.

The symbiotic relationship between dairy cows and rumen microbiota enables them to utilize non-protein nitrogen (urea and ammonia) and enhance the biological value of dietary protein through rumen microbial protein synthesis [30,31,32]. Ammonia is absorbed through the walls of the rumen and enters the liver, where it is metabolized into urea and circulated back into the rumen or transported to the kidneys [33]. Urease hydrolyzes urea into ammonia, which can be excreted from the stool or absorbed again by the rumen wall and subsequently metabolized into urea in the liver [33]. However, this repeated production of urea can be an energy-consuming process [34]. The faster the microbial energy source, the higher the production of VFA. Qiao et al. evaluated the effect of dietary energy nitrogen synchronization on rumen fermentation of Holstein dairy cows and found that TVFA increased linearly with an increase in energy nitrogen release synchronization [35]. Similar results were found in this study, with TVFA in the treatment group significantly higher than that in the control group. Carbohydrates that are easily fermented may increase acetic acid production, while proteins and non-structural carbohydrates may increase propionic acid production. The simultaneous release of energy and nitrogen optimizes the rumen fermentation process and promotes the production of specific VFA. It was found that propionic and butyric acid concentrations increased significantly when the release rates of energy and nitrogen sources were matched [29].

According to the results of alpha diversity, compared with the CON, the richness of microflora in all treatment groups decreased, but the community uniformity increased, which may be related to the balance of energy and nitrogen in the rumen, and may promote the growth of the dominant microbial flora in the rumen. It has been shown that different levels of synchronization of energy and nitrogen release significantly alter the relative abundance of rumen microorganisms. For example, the relative abundance of certain bacterial phyla, such as Bacteroidetes and Firmicutes, increased with increasing levels of energy nitrogen synchronization [29].

Studies have shown that Prevotellaceae can utilize carbohydrates and proteins in feed, and the main fermentation products are acetic acid, propionic acid, and succinic acid. Propionic acid is the substrate of gluconeogenesis, which can activate gluconeogenesis and maintain energy stability [36,37]. Consistent with the results of this experiment, acetic acid and propionic acid concentrations also increased with the relative abundance of Prevotellaceae. At the family level, Prevotellaceae was dominant, and the relative abundance of Prevotellaceae in the MMU was higher than that in the other groups, which was consistent with that of Bacteroidetes at the phylum level. Lachnospiraceae is the most abundant Firmicutes in the rumen environment, and it belongs to Firmicutes with Ruminococcaceae. Most species positively correlated with carbohydrate metabolism belong to Lachnospiraceae, many of whose members have cellulose-breaking activity and are associated with the production of butyric acid, which is an important energy source for rumen epithelial cells, playing a role in regulating rumen barrier function and potentially affecting feed conversion [38,39]. In this experiment, compared with the CON, the relative abundance of Lachnospiraceae and Ruminococcaceae in all treatment groups increased, which may be the reason for the increase in the relative abundance of Firmicutes in all treatment groups. Moreover, we also found that the concentration of butyric acid in the rumen fermentation index showed an upward trend. This is consistent with the results of Mu et al. [40]

At the genus level, Prevotella can utilize fermentable carbohydrates, including molasses and soluble fibre from liquid raw materials [41]. This is consistent with the results of this study. Studies have shown that Ruminococcus and Prevotella are bacteria that play a major role in decomposing feed fibre and starch in the rumen. The relative abundance of the two bacteria genera in all treatment groups increased compared with the other treatment groups, which may be because the experiment changed the energy and nitrogen release synchronization of raw materials in the diet formula, resulting in an increase in the relative abundance of the two bacteria genera. The abundance of norank_f__Bacteroidales_RF16_group bacteria in the rumen of yaks with different feed types was found to be different; the rumen microbial abundance of yaks fed roughage was higher than that of those fed concentrate [42]. In this experiment, the bacterial abundance of the RF16_group gradually increased, possibly because with the increase in molasses, it matched the rate of nitrogen release from slow-release urea, promoting microbial growth and thus increasing the abundance of this bacterium. Succiniclasticum had the highest abundance in the CON and decreased in all treatment groups, while the MMU had the lowest abundance. Carbohydrate molasses, which were easier to ferment than starch, were added to each treatment group, so the relative abundance of Succiniclasticum decreased.

The results showed that the CON and MMU had significant differences in the structure of rumen bacterial flora. In this experiment, the abundance of cyanobacteria in the CON was significantly higher than that in the other treatment groups, and the study showed that most bacteria of cyanobacteria were negatively correlated with total volatile fatty acids, acetic acid, propionic acid, and butyric acid production in the rumen [43]. This was consistent with the results of this experiment, and the yields of total volatile fatty acids, acetic acid, propionic acid, and butyric acid in the experimental group were higher than those in the CON. Fibrobacterita can greatly enhance the synergistic effect of microorganisms in the process of lignocellulose degradation in the rumen environment. It is well known that metabolite malnutrition can enhance cooperation between microbial communities, thus promoting the stability of the entire microbiota [44]. Patescibacteria, formerly known as candidate phyla radiation (CPR), are a class of genes that have lost major metabolic pathways such as amino acids, nucleotides, fatty acids, and cofactors for de novo synthesis, and are dependent on other microorganisms. Therefore, it has been considered a phylum of malnutrition metabolism [44].

In this experiment, the relative abundance of Fibrobacterita and Patescibacteria in the treatment group was lower than that in the CON, indicating that the diet in the treatment group was more in line with the digestion and absorption of rumen microflora than that in the CON, further indicating the necessity of the synchronization of energy nitrogen release in ruminant nutrition. Colidextribacter and Selenomonas were significantly enriched in the MMU. Colidextribacter and Selenomonas play important roles in rumen carbohydrate metabolism; Colidextribacter is able to break down complex carbohydrates, such as cellulose, and convert them into simple sugars, which provide more energy for the animal and help maintain a healthy rumen by participating in the synthesis of short-chain fatty acids [45]. Selenomonas, on the other hand, utilizes lactic acid as a source of energy, and converts it into propionic acid and other metabolites, thereby reducing the concentration of lactic acid in the rumen and preventing acidosis caused by lactic acid accumulation. In addition, Selenomonas can degrade complex carbohydrates into simple sugars and short-chain fatty acids, further contributing to energy metabolism and the stability of the rumen environment [46]. It has been reported that the relative abundance of Colidextribacter is positively correlated with the concentration of propionic acid [47]. The study found that adjusting the level of Colidextribacter in the goat model increased the concentration of butyric acid in the treatment group [48], which was similar to the results of this experiment. Colidextribacter was significantly enriched in the MUN group, and Colidextribacter produced SCFA (mainly butyrate). Selenomonas can ferment lactic acid to produce propionic acid, which plays a key role in the utilization of lactic acid and the reduction in methane production in the rumen [49]. The increase in the relative abundance of the two strains resulted in an increase in propionic acid concentration, which corresponded with the change in rumen fermentation parameters in the MMU. Lyons et al. found that the relative abundance of Acetobacter in the rumen of cows in mid-lactation was significantly higher than that in late lactation [50]. Acetobacter can produce acetic acid by oxidizing sugar, and most of its members are aerobic bacteria, which may be the reason for the increase in acetic acid concentration in the HMU.

5. Conclusions

The present study was an exploratory study aimed at preliminarily investigating the effects of different levels of molasses and slow-release urea on rumen microbial activity, fermentation, and ammonia concentration in Holstein fattening cows. In conclusion, the present study showed that the use of 3% slow-release urea and 4% molasses as partial concentrate substitutes could effectively improve the growth performance and rumen fermentation of Holstein fattening bulls, affect the rumen microbiota, and improve the production performance. It was found that the appropriate dose of molasses and slow-release urea could improve the feed utilisation efficiency of Holstein fattening bulls by regulating the rumen microbial community structure, which in turn could promote the growth performance of Holstein fattening bulls. Compared with the CON, the cattle fed with the combination of 4% molasses and 3% slow-release urea could reduce the feeding cost by RMB 1.62 per day, and increase the farming benefit by RMB 7.19 per cow per day. It provides a cost-effective strategy to improve production performance and reduce feeding costs. Different sources of molasses and preparation processes may lead to differences in the results, which to some extent affects the generalizability of the findings. In addition, the experimental period of the present study was relatively short, and the long-term effects of the combination of molasses and slow-release urea and its potential cumulative effects need to be further explored.