Abstract Background The addition of wine lees to diets can make up for the deficiencies caused by traditional forages in beef cattle farming. However, the effects of different wine lees ratios on average daily weight, gastrointestinal microbial community structure and metabolites in Guanling crossbred cattle have been rarely studied. This study assessed the effects of feeds containing wine lees on weight gain, gastrointestinal microbial community structure, and metabolites in Guanling crossbred cattle and elucidated the metabolic responses induced by wine lees. Eighteen cows were randomly assigned to receive fed concentrate (C group), feed containing 15% wine lees (group A), or feed containing 30% wine lees (group B) for 60 days. Results The average daily weight gain of group A and group B increased by 76.75% and 57.65%, respectively, compared with group C. Microbial community analysis showed that wine lees increased the abundance of Prevotella_1 in the rumen, decreased the abundance of Ruminococcaceae UCG 011 and Lachnospiraceae_FCS020_group in the rumen, and increased the abundance of Tyzzerella_4, Family_Xlll_AD3011_group, Granulicella, and Eisenbergiella in the cecum. Metabolomics analyses showed that wine lees decreased the concentrations of indole-3-ethanol in the rumen, and complexity cecal metabolism. Notably, linoleic acid metabolism was significantly enriched in both the rumen and cecum. Mantel test analyses indicated that the adverse effects of WL were reduced by stimulating the metabolism of linoleic acid, α-linolenic acid, and tryptophan, and these changes were mediated by intestinal microorganisms. The Guanling cattle cecum was enriched for several unfavorable metabolic pathways when wine lees concentrations reached 30%, which increased the likelihood of intestinal lesions. Conclusion This study shows that WL supplementation alters gut microbiota and metabolic pathways, improving cattle growth and health. Moderate WL levels (15%) enhance gut health and beneficial pathways (e.g., linoleic and alpha-linolenic acid metabolism). However, higher WL inclusion (30%) may activate adverse pathways, raising the risk of intestinal damage. To maximize benefits and minimize risks, WL levels should be carefully managed. Supplementary Information The online version contains supplementary material available at 10.1186/s12866-024-03583-z. Keywords: Gut bacteria, White wine lees, Guanling crossbred cattle, 16S rDNA sequencing, LC–MS Highlights • Feeding wine lees(WL) increased average daily weight gain of Guanling crossbred cattle • Changes in the gut flora of Guanling crossbred cattle induced by feeding WL. • Excess WL (30%) increased the likelihood of intestinal lesions in Guanling crossbred cattle. • Guanling crossbred cattle are stimulated to reduce the adverse effects of WL by regulating the metabolism of linoleic acid, alpha-linolenic acid and tryptophan. Supplementary Information The online version contains supplementary material available at 10.1186/s12866-024-03583-z. Introduction Livestock plays a key role in food production in low- and middle-income countries by promoting economic growth and reducing poverty [[40]1]. Since 1980, the global demand for livestock products has more than doubled in response to the rapid growth of the global population and the substantial improvement in people’s quality of life [[41]2]. Moreover, traditional forages no longer meet livestock production demands, and by-products must be effectively utilized to improve feed utilization. White wine lees (WL), made from fermented sorghum and corn, are inexpensive and rich in fat, proteins, fibers, vitamins, trace elements, and other nutrients [[42]3]. Compared to traditional feeds such as hay and maize, it has a higher nutritional value. In particular, the high content of polyphenolic compounds in WL has significant antioxidant and anti-inflammatory properties [[43]4], which can enhance the immunity and intestinal health of cattle. Furthermore, the economic and environmental advantages of using WL as a feed additive are worth mentioning as it reduces waste management costs and lowers feed expenses. However, the high acidity and water content of WL increase spoilage and shorten shelf life, and the disposal of this product pollutes the environment and wastes resources. Therefore, the rationale for using this product is challenging. WL as feed improves animal growth performance, health and immunity, and disease resistance [[44]5, [45]6]. Moreover, WL polyphenols reduce colitis in mice, and this effect is mediated by intestinal microorganisms [[46]4]. However, some WL components impair health by changing the gut microbiome [[47]7]. Intestinal health affects livestock growth, and our previous study demonstrated the positive effects of WL feed on intestinal health in Guanling cattle [[48]8]. However, excessive WL intake produced unfavorable metabolic changes in beef cattle. In addition, the effect of WL feed on the intestinal health of beef cattle and the mechanisms underlying resistance to the adverse effects of WL are unclear. Elucidating such mechanisms is essential to ensure the rational use of WL as feed. Microbiome analysis assesses the relationship between animals and their microbial symbionts and identifies potential metabolic biomarkers in microbial communities [[49]9]. This study assessed the effects of feed containing different concentrations of fermented WL on Guanling crossbred cattle. Changes in microbial community structure and gastrointestinal metabolism in WL-fed cattle were investigated by 16S rDNA sequencing and liquid chromatography-mass spectrometry (LC–MS) to identify metabolic responses to WL. Materials and methods WL preparation WL was obtained from Mao Tai Distillery (Renhuai, Guizhou Province). Microbial ferments were purchased from Yijiayi Biological Engineering Co., Ltd. (Batch number: Pre-feeding Tim. (2015) 13524). Considering the nutritional intake of Guanling crossbred cattle, WL was pretreated in this study. 1 tonne of WL was pretreated in the following way: 92% of Moutai wine dregs, 3% of maize flour, rapeseed meal and 2% of wheat bran were mixed well, and then added to an activated microbial fermenter (1 kg of brown sugar and 200 g of microbial preparation in 10 L of water). Fermentation was carried out for 3–7 d before feeding. Experimental animals Eighteen female Guanling crossbred (Guanling × Simmental) cows (age, 18 months; weight, 250 ± 25 kg) were obtained from the Guizhou Provincial Yellow Cattle Group. The animals were free from infectious diseases (brucellosis, foot-and-mouth disease virus types A and O), trauma, depression, and loss of appetite. Animal grouping and treatment The animals were housed in pens and randomly assigned to receive feed concentrate alone (control group [CK]), feed containing 15% WL as concentrate replacement (group A), and feed containing 30% WL as concentrate replacement [group B]), with six animals per group. All animals had ad libitum access to forage and water. The feeds were formulated according to the Chinese beef cattle feeding industry standard (NY/T815 -2004) and the formulations are shown in Table [50]1. Beef cattle were fed daily at 09:00 and 16:30 h during the trial period, which lasted 75 d (15 d for the pre-feeding period and 60 d for the official feeding period). Hygiene and daily management according to routine methods. This work utilized the UNIH Guide for the Care and Use of Laboratory Animals, was performed according to the Declaration of Helsinki and was approved by the Guizhou University Review Committee for the Use of Human or Animal Subjects (Ethics number: EAE-GZU-2020-E018). Table 1. Composition and nutrient content of the feed Item Groups CK A B Composition (%) Royal bamboo grass 60.00 60.00 60.00 Corn 20.00 10.00 0.00 Soybean meal 6.00 3.00 0.00 Rapeseed cake 4.00 2.00 0.00 Wheat bran 6.00 6.00 6.00 Fermented wine lees 0.00 15.00 30.00 Premix 2.00 2.00 2.00 Stone powder 1.45 1.45 1.45 Salt 0.05 0.05 0.05 Calcium bisulfate 0.42 0.42 0.42 Sodium sulfate 0.08 0.08 0.08 Total 100.00 100.00 100.00 Nutrient content Dry matter, % 60.58 60.68 60.20 Digestive energy, MJ/kg 13.10 13.09 13.08 Crude protein, % 13.15 13.65 14.32 Neutral detergent fiber, % 56.26 53.24 46.32 Acid detergent fiber, % 35.57 34.09 27.58 Fat, % 2.67 2.89 3.13 Total phosphorus, % 0.46 0.46 0.47 Calcium, % 0.82 0.83 0.83 [51]Open in a new tab CK, fed feed without wine lees; A, fed feed with 15% wine lees; B, fed feed with 30% wine lees At the end of the experiment, three cows were randomly selected from each group and transferred to the slaughterhouse for electroshock slaughter; cows to be executed were prohibited from feeding for 12 h before slaughter. The rumen and cecum contents of cows in each group were collected separately in 50 mL sterile test tubes. The 18 samples collected were rapidly frozen in liquid nitrogen for 12 h and subsequently transferred to -80 °C for storage until microbiological and metabolomics analyses were performed. Microbiological analysis Genomic DNA was extracted using the DNeasy PowerSoil kit. The V3-V4 hypervariable region of the 16S rDNA gene was amplified using primers 343F (5'-TACGGRAGGCAGCAG-3') and 798R (5'-AGGGTATCTAATCCT-3'). Polymerase chain reaction (PCR) was performed using Tks Gflex DNA Polymerase (580BR10905, Bio-Rad). The reaction mixture contained 15 μL of 2 × Gflex PCR buffer, 1 μL of each primer (5 pmol/μL), 1 μL of template DNA, 0.6 μL of Tks Gflex DNA Polymerase (1.25 U/μL), and 11.4 μL of nuclease-free water, in a total volume of 30 μL. The amplification conditions consisted of a denaturation cycle at 94 °C for 5 min, followed by 26 cycles at 94 °C for 30 s, 56 °C for 30 s, and 72 °C for 20 s, and a final extension step at 72 °C for 5 min. PCR amplicons were sequenced by Ouyi Biomedical Technology Co. Raw sequencing data were stored in FASTQ format, and sequences with base quality below 20 or lengths shorter than 50 bp were removed using Trimmomatic (version 0.35). Overlapping paired-end reads were merged via Flash (version 1.2.11). High-quality merged sequences were retained for further analysis, with criteria including the absence of ambiguous base “N”, fewer than 8 single-base repeats, and a read length exceeding 200 bp, processed via the split_libraries tool in QIIME (version 1.8.0). Chimeric sequences were detected and eliminated using UCHIME (version 2.4.2). The remaining sequences were clustered into operational taxonomic units (OTUs) based on a 97% similarity threshold. Taxonomic assignment was performed using the RDP classifier (version 2.2) with a confidence threshold of 0.7–1. Alpha diversity was used to assess bacterial richness and diversity. Beta diversity was calculated using Bray–Curtis distances and visualized by principal coordinate analysis (PCoA) in QIIME (version 1.8.0) and R (version 2.15.3). Genus-level linear discriminant analysis effect size (LEfSe) was conducted using the Kruskal–Wallis rank sum test, with significance set at p < 0.05 and a log-linear discriminant analysis (LDA) score threshold of 2.0. Metabolomics analysis Metabolic profiling was performed using an Acquity UPLC I-Class system coupled to a VION IMS Q-TOF high-resolution mass spectrometer (Waters Corp.) and an Acquity UPLC BEH C[18] column (100 mm × 2.1 mm, 1.7 μm). The following chromatographic conditions were used: column temperature, 45 °C; eluent A, 0.1% formic acid in water; eluent B, 0.1% formic acid in acetonitrile; flow rate, 0.4 mL/min; injection volume, 1 μL. In addition, the following spectrometric parameters were used: ion modes, positive and negative; ion source, electrospray; electrospray capillary voltage, 2.5 kV; injection voltage, 40 V; collision voltage, 4 eV; ion source temperature, 115 °C; desolvation temperature, 450 °C; carrier gas flow rate, 900 L/h; scan range, 50–1000 amu; scan time, 0.2 s; time between scans, 0.02 s. Statistical and correlation analysis The correlation between taxonomic groups (phyla and genera) and metabolites was analyzed by Pearson correlation analysis and analysis of variance (variable influence on projection (VIP) > 1 and p < 0.05). All statistical analyses were performed using SPSS Statistics version 22.0 (IBM Corporation, Armonk, NY, USA) for Windows. Continuous variables were expressed as means ± standard deviations, and p-values of less than 0.05 were considered statistically significant. Results Daily weight gain (DWG) Animals were weighed every 30 days. After 60 days of feeding, the average DWG of groups A and B increased by 76.75% and 57.65%, respectively, compared with CK. The average DWG in group B increased by 50.42% from the first to the second month (Table [52]2). WL significantly increased DWG; however, weight gain was not linearly proportional to the amount of WL added to the feed. Table 2. Average daily weight gain Item Percent of wine lees in the feed Average daily weight gain Total average daily weight gain SEM First month Second month CK 0 0.61 0.70 0.65 0.14 A 15 1.42 0.92 1.16 0.50 B 30 1.25 0.83 1.03 0.22 [53]Open in a new tab CK, fed feed without wine lees; A, fed feed with 15% wine lees; B, fed feed with 30% wine lees Sequencing data quality control The effect of WL on gastrointestinal microorganisms was assessed by the number of OTUs. WL nonlinearly decreased the number of unique OTUs in the rumen and increased the number of unique OTUs in the cecum (Fig. [54]1a, b). The number of OTUs was higher in the rumen than in the cecum. Fig. 1. [55]Fig. 1 [56]Open in a new tab Intestinal microbial composition of Guanling crossbred cattle (a) comparative OUT analysis of ruminal colonies; b comparative OUT analysis of cecum colonies; c PCoA analysis of ruminal colonies; d PCoA analysis of cecum colonies. CK, fed feed without wine lees; A, fed feed with 15% wine lees; B, fed feed with 30% wine lees Analysis of alpha diversity indexes Microbial alpha diversity in the rumen and cecum is shown in Table [57]3. Good’s coverage index was close to 1, indicating that sequencing covered most bacterial taxa. Alpha diversity was similar across treatments in the rumen and cecum. Nonetheless, the observed species in the rumen increased with increasing lees additions. Table 3. Alpha diversity indices Groups Observed species Chao1 Shannon Simpson Good’s coverage PD whole tree Rumen CK 6448.20 9435.46 9.71 0.97 0.95 237.09 A 5906.93 8735.00 9.61 0.99 0.96 222.51 B 6989.00 10,381.11 10.52 1.00 0.95 251.98 Cecum CK 6989.00 7499.14 9.89 1.00 0.96 192.34 A 5187.63 7783.80 9.83 1.00 0.96 193.35 B 4901.57 7444.42 9.52 0.99 0.96 186.98 [58]Open in a new tab Sequences were selected randomly, and alpha diversity indices were calculated at the lowest sequencing depth Analysis of bacterial community composition The effects of WL on bacterial community composition were evaluated by principal component analysis (PCA). The results indicated that WL altered bacterial communities in the rumen and cecum (Fig. [59]1c, d). Changes in these communities were more pronounced in the rumen. Amplicon sequences were assigned to taxonomic groups (phylum and genus) using the Ribosome Database Project classifier and the SILVA database. At the phylum level, WL increased the relative abundance of Bacteroidetes by 7.43–14.01% (Fig. [60]2a). The predominant phylum in the cecum was Firmicutes, and WL increased its relative abundance (Fig. [61]2c). Fig. 2. [62]Fig. 2 [63]Open in a new tab Structure of the intestinal bacterial community in Guanling crossbred cattle (a) relative abundance of the top 10 bacterial phyla in the rumen; b relative abundance of the top 10 bacterial genera in the rumen; c relative abundance of the top 10 bacterial phyla in the cecum;d relative abundance of the top 10 bacterial genera in the cecum; and LefSe analyses of rumen bacteria (e) and cecum bacteria (f) The predominant genera in the rumen were Prevotella_1 and Rikenellaceae_RC9_gut_group. Ruminococcaceae_UCG-005 and Ruminococcaceae_UCG-010 were the predominant genera in the cecum, and their abundance was not affected by WL. In contrast, WL increased the relative abundance of Prevotella_1 (P < 0.05) (Fig. [64]2b). Linear discriminant analysis effect size (LefSe) analysis showed that 30% WL decreased the relative abundance of Ruminococcaceae UCG 011 and Lachnospiraceae_FCS020_group (LAD score > 3, Fig. [65]2e) and increased the relative abundance of Family_Xlll_AD3011_group_g, Tyzzerella_ 4, Eisenbergiella, and Granulicella in the cecum (LAD score > 3, Fig. [66]2f). These results suggest that WL changed gut microbial abundance in Guanling crossbred cattle. WL affects intestinal metabolism The effects of WL on intestinal metabolism were assessed by calculating the number of unique metabolites. The results showed that WL increased the number of unique metabolites in the rumen and cecum in a dose-dependent manner (Fig. [67]3a). In addition, PLS-DA score plots showed that the samples from each treatment tended to cluster together (Fig. [68]2 c, d). This result suggests that WL changed the intestinal metabolic profile. Fig. 3. [69]Fig. 3 [70]Open in a new tab (a) Venn diagram of ruminal metabolites in Guanling crossbred cattle; b Venn diagram of cecum metabolites in Guanling crossbred cattle; c PLS-DA analysis of ruminal metabolites in Guanling crossbred cattle; d PLS-DA analysis of cecum metabolites in Guanling crossbred cattle. CK, fed feed without wine lees; A, fed feed with 15% wine lees; B, fed feed with 30% wine lees Analysis of differential metabolites The metabolites with VIP >1 in the first principal component of the OPLS-DA model and p < 0.05 in the t-test were considered differentially expressed. The effects of WL on the number of differential metabolites are shown in Fig. [71]4 a-d. The number of differential metabolites and the concentration of indole-3-ethanol decreased in the rumen as WL concentration increased. In contrast, the number of differential metabolites increased from 49 to 75 in the cecum during the experimental period. Fig. 4. [72]Fig. 4 [73]Open in a new tab Analysis of intestinal metabolites in Guanling crossbred cattle (a) volcano plots of differential metabolites between CK and A in the rumen; b volcano plots of differential metabolites between CK and B treatments in the rumen; c volcano plots of differential metabolites between CK and A in the cecum; d volcano plots of differential metabolites between CK and B in the cecum; e analysis of enrichment of KEGG metabolic pathway. Volcano plots are labeled only for different metabolites enriched to the KEGG metabolic pathway, with blue indicating down-regulation and red indicating up-regulation * denotes significant differences. CK, fed feed without wine lees; A, fed feed with 23 15% wine lees; B, fed feed with 30% wine lees Pathway enrichment analysis of differential metabolites was performed using the KEGG database (Fig. [74]4e). WL significantly enriched intestinal metabolic pathways, including the metabolism of alpha-linolenic acid, linoleic acid, tryptophan, and choline. Among them, 13 pathways were modulated in the cecum of treated animals, indicating that WL had the greatest effect in the cecum. Correlation of microorganisms with metabolites We hypothesized that microorganisms mediate the effect of WL on intestinal metabolism. Therefore, we performed a Mantle test with key bacteria identified in the Lefse analysis as causing changes in gut microbial composition (LAD score > 3) and with differential metabolites in the KEGG metabolic pathway. The results showed that the abundance of Lachnospiraceae_FCS020_group was positively correlated with indole-3-ethanol and 3-alpha-androstanediol glucuronide concentrations in the rumen (Fig. [75]5a). Three bacterial taxa—Tyzzerella, Granulicella, and Eisenbergiella—were significantly correlated with several metabolites in the cecum (Fig. [76]5b). Fig. 5. [77]Fig. 5 [78]Open in a new tab Mantal test analysis of rumen bacteria (a) and cecum bacteria (b) with differential Discussion Wine production is increasing in China, and the use of by-products, including WL, can reduce waste and environmental pollution. WL increases beef cattle performance [[79]10]. In the present study, WL increased the average DWG in Guanling crossbred cattle. Average DWG is a good indicator of animal growth and development and is influenced by intestinal metabolism and flora [[80]11]. Therefore, the DWG changes may be due to WL-mediated changes in the intestinal microbial community. Analysis of the intestinal bacterial community composition using 16S rDNA sequencing revealed that wine lees (WL) supplementation did not significantly impact overall alpha diversity of gut bacteria. This observation aligns with previous research [[81]8], suggesting that the introduction of WL does not disturb the established microbial diversity, a key factor in maintaining gut homeostasis [[82]12]. The maintenance of stable alpha diversity indicates that the microbial ecosystem remains balanced, which is crucial for supporting essential functions such as nutrient absorption, immune response, and overall health of the host organism. Thus, WL supplementation appears to support this microbial equilibrium, potentially enhancing gut health without causing microbial dysbiosis. Bacteria, which comprise over 95% of the gastrointestinal microorganisms, play a pivotal role in nutrient digestion and absorption in livestock [[83]13]. For instance, gut bacteria ferment more than half of the dry matter and crude fiber in livestock feed, producing metabolites such as ATP, volatile fatty acids, B vitamins, and proteins that are vital for both microbial and host metabolism. These metabolites not only serve as energy sources for the host but also as substrates for other beneficial microbial processes [[84]14]. The composition and function of these microbial communities are significantly influenced by the type of feed provided. Consistent with this understanding, our study demonstrated that WL supplementation leads to notable shifts in the relative abundance of specific bacterial taxa within the intestines of Guanling cattle (Fig. [85]2). These shifts likely represent an adaptive response to optimize digestion and nutrient utilization, rather than a disruption of microbial balance. The most abundant phyla in the rumen and cecum were Firmicutes and Bacteroidetes, consistent with previous findings [[86]15]. With the addition of WL, the relative abundance of Bacteroidetes in the rumen gradually increased, whereas the relative abundance of Firmicutes showed a decreasing trend. The increase in Bacteroidetes and the simultaneous decrease in Firmicutes imply an overall increase in the Bacteroidetes to Firmicutes (B/F) ratio. A higher B/F ratio is generally associated with improved gut health, as Bacteroidetes are involved in the degradation of complex carbohydrates, leading to the production of beneficial short-chain fatty acids (SCFAs) [[87]16]. These SCFAs play a crucial role in maintaining the integrity of the gut lining, modulating immune responses, and providing energy to host cells. However, in the cecum the opposite result to that of the rumen is shown, variation that may be related to the fact that the rumen and the cecum have different functions in the digestive system of ruminants. The rumen is mainly responsible for the initial fermentation and cellulose breakdown, whereas the cecum is the site of further fiber fermentation and nutrient absorption. WL may facilitate the degradation of complex carbohydrates by Bacteroidetes in the rumen, but in the cecum, the predominance of Firmicutes may be due to their important role in the utilization of fermentation end-products. Overall, the effect of WL on the B/F ratio underscores its potential as a dietary supplement to improve gut health in Guanling crossbred cattle. However, it is important to balance WL supplementation to avoid possible adverse effects, such as those observed at higher concentrations. At a genus level, WL increased the relative abundance of Prevotella_1 in the rumen but not in the cecum. Prevotella is the most abundant genus in the rumen and degrades and utilizes plant fibers in the rumen of dairy cows [[88]17]. However, changes in the abundance of dominant bacterial genera do not explain interspecies differences and the effects of these characteristics. Therefore, we used Lefse analyses to account for interspecific variation and the effects of these traits as a means of identifying the main bacteria influencing changes in the flora [[89]18]. This analysis showed that it was mainly Ruminococcaceae UCG 011 and Lachnospiraceae_FCS020_group that caused the changes in rumen gut bacterial flora. Ruminococcaceae UCG 011 and Lachnospiraceae_FCS020_group all belong to the order Clostridia, which are able to metabolize tryptophan [[90]19]. Tryptophan, an essential amino acid, plays a critical role in reducing animal stress and enhancing growth performance by supporting feed utilization [[91]20]. Moreover, tryptophan-derived metabolites, such as indole and its derivatives, are vital for maintaining gut barrier integrity and regulating inflammatory responses [[92]21–[93]23]. Under WL supplementation (including 30%), there was a significant reduction in these tryptophan-metabolizing bacteria, potentially leading to a decreased production of protective metabolites like indole-3-ethanol. These metabolites are crucial for modulating gut permeability and suppressing pro-inflammatory cytokine production. A reduction in their levels may compromise the intestinal barrier, increasing the risk of inflammation and disorders like leaky gut syndrome. The cecum of ruminants is an essential site for food digestion and absorption. Tyzzerella 4, Granulicella and Eisenbergiella are the main bacteria responsible for changes in the intestinal flora of the cecum. can improve intestinal function by converting complex polysaccharides into short-chain fatty acids, including acetic, butyric, and propionic acids, and increasing resistance to colonization [[94]24, [95]25]. The increase in the relative abundance of Tyzzerella_4, Granulicella, and Eisenbergiella by WL in the cecum confirms the positive effect of this product on gut health. However, 30% WL slightly increased the relative abundance of Family_Xlll_AD3011_group. family_Xlll_AD3011_group was positively correlated with fasting insulin and blood glucose [[96]26]. Furthermore, several studies suggest that Family_XIII_AD3011_group may serve as a potential marker of inflammation in dairy cows, with its increased abundance potentially triggering inflammatory responses [[97]27]. WL levels should be carefully regulated to avoid potential long-term health risks. Metabolomics analysis showed that WL modulated the expression of several intestinal metabolites, including lipids, phenylpropanoids, polyketides, and organoheterocyclic compounds. These metabolites are essential for microbial growth [[98]28]. For instance, lipids serve as vital energy sources and are integral to the formation of cell membranes in bacteria. Phenylpropanoids and polyketides, on the other hand, are involved in various biosynthetic pathways that produce secondary metabolites, which can influence microbial interactions and competition within the gut. Organoheterocyclic compounds often act as signaling molecules or intermediates in essential metabolic processes, contributing to the overall metabolic network that supports bacterial growth and stability. The modulation of these metabolites by WL suggests that this feed additive can influence the microbial ecosystem by altering the availability of these crucial compounds, thereby shaping the gut microbiome's structure and function. Indole-3-ethanol and alpha-linolenic acid were involved in tryptophan metabolism (treatment B) and α-linolenic acid metabolism (treatment A, Fig. [99]4e). Tryptophan metabolism produces indole and indole derivatives, which reduce intestinal inflammation [[100]21–[101]23]. WL decreased indole-3-ethanol levels probably by decreasing the abundance of tryptophan-metabolizing bacteria, impairing gut health. These findings were corroborated by the results of the Mantel test (Fig. [102]5a). WL at 15% increased the concentrations of alpha-linolenic acid, an essential fatty acid that modulates inflammatory signaling [[103]29]. In addition, WL increased the levels of other lipids, which provide fatty acids and energy and regulate intestinal metabolism and immunity [[104]30]. One possible explanation for this phenomenon is that the animals are attempting to resist the effects of unfavorable factors. However, this self-regulation decreases when the concentration of WL reaches 30%. WL increased the number of several metabolites in the cecum, including lipids, phenylpropanoids, polyketides, organoheterocyclic compounds, organic oxygen compounds, alkaloids and derivatives, organic acids and derivatives, nucleosides, and nucleotides. These metabolites belong to different pathways. Differential metabolites that were shared and significantly enriched into metabolic pathways in the A and B treatments were 9,12,13-TriHOME,17beta-Estradiol 3-(beta-D-glucuronide) and 3-(3-Indolyl)-2-oxopropanoic acid. Among these, linoleic acid-derived 9,12,13-TriHOME reduces inflammation, regulates metabolism, slows aging, prevents fatigue, and stimulates memory phenotype CD8^+ T cells [[105]31]; 3-(3-indolyl)-2-oxopropanoic acid, upregulated in the cecum, improves tryptophan metabolism and reduces intestinal inflammation. The upregulation of 3-(3-indolyl)-2-oxopropanoic acid and 9,12,13-TriHOME by WL may be due to the increased abundance of Granulicella and Eisenbergiella (Fig. [106]5b). WL also upregulated tyrosine. In addition, the upregulation of LysoPC (15:0) was associated with the increased abundance of Eisenbergiella in the intestine of animals treated with 15% WL. LysoPC (15:0) is a marker of metabolic syndrome, and LysoPC (15:0) levels are positively associated with the risk of overweight, obesity, dyslipidemia, hyperuricemia, hyperinsulinism, and high insulin resistance (assessed using the homeostasis model assessment of insulin resistance) [[107]32]. Further, LysoPC (15:0) is positively correlated with 9,12,13-TriHOME. We hypothesize that 9,12,13-TriHOME levels are increased to counteract the effects of LysoPC (15:0). This finding explains the absence of abnormalities in beef cattle in this study. The results of Mantel test showed that 17-β-estradiol 3-beta-D-glucuronide concentrations were positively correlated with the abundance of Tyzzerella_4, Eisenbergiella, and Family_Xlll_AD3011_group; nonetheless, it was significantly down-regulated in A, B treatments compared to CK. which might be because of the 9,12,13-TriHOME modulated the expression of 17–estradiol 3-D-glucuronide (Fig. [108]5b). Prostaglandin F2a as a metabolite of estrogens might be the same. In addition, 30% WL upregulated L-tyrosine and enriched several metabolic pathways, including melanogenesis, cocaine addiction, amphetamine addiction, alcoholism, and Parkinson’s disease. These pathways are associated with pathological states [[109]8], implying that excessive amounts of WL are unsuitable as feed for Guanling cattle. 9,12,13-TriHOME, which improves health [[110]33], with up-regulation observed only in the A treatment, and this was supported by the down-regulation of Resveratrol with therapeutic potential [[111]34] in treatment B. Therefore, the content of WL fed to Guanling crossbred cattle should not be too high and should be controlled at less than 30 percent. Overall, metabolomic analysis revealed that several key metabolites were modulated by different WL inclusion rates, providing insights into their broader physiological effects on cattle health. For instance, α-linolenic acid, which was significantly elevated in the 15% WL group, is known for its anti-inflammatory properties and its role in maintaining cell membrane integrity. This fatty acid also contributes to immune regulation and helps reduce intestinal inflammation, suggesting that moderate WL inclusion can promote overall gut health by enhancing these protective mechanisms. On the other hand, in the 30% WL group, we observed an upregulation of LysoPC (15:0), a lysophosphatidylcholine associated with metabolic syndromes such as insulin resistance and cardiovascular risks. This finding indicates that high WL concentrations may impose metabolic stress on the cattle, disrupting lipid metabolism and potentially leading to long-term health issues. Furthermore, the reduction in tryptophan metabolites, such as indole-3-ethanol, in the 30% WL group is concerning, as these metabolites play a crucial role in maintaining gut barrier function and regulating inflammation. A decrease in these protective metabolites may compromise intestinal integrity, making the cattle more susceptible to inflammation and gut disorders. WL supplementation significantly altered the intestinal bacterial community composition and affected intestinal metabolic functions in Guanling crossbred cattle. While WL promotes beneficial pathways related to linoleic acid, α-linolenic acid, and tryptophan metabolism that may help mitigate some negative effects, our findings demonstrate a threshold effect, with adverse health impacts observed when WL concentration reaches 30%. These insights offer a novel perspective on optimizing WL use in cattle feed, balancing the benefits of enhanced nutrient metabolism with the risk of potential gut health disruption. Farmers are advised to limit the addition of WL to about 15%. At this concentration, WL significantly increased daily weight gain and improved gut health while avoiding the adverse effects observed at higher concentrations (30%). Conclusions WL supplementation in Guanling crossbred cattle altered specific gut microbial taxa without significantly affecting overall α-diversity. Enrichment of metabolic pathways associated with linoleic acid, α-linolenic acid and tryptophan suggests that WL may enhance gut health and nutrient utilization. However, high levels of WL (30%) may activate pathways associated with intestinal damage, suggesting that the dose needs to be adjusted carefully. The implication is that WL could be used as a sustainable feed additive to promote growth and health of beef cattle while reducing environmental waste. Future research should focus on the long-term impacts of WL supplementation, the underlying mechanisms of microbial-host interactions, and the broader applicability of WL across different breeds and production systems to optimize its use in livestock nutrition. Supplementary Information [112]Supplementary Material 1.^ (15.8KB, docx) Acknowledgements