Introduction

Lipid metabolism dysfunction in late-stage laying hens is a significant contributor to the decline in both laying rates and egg quality, leading to substantial economic losses in the egg farming industry [1]. In poultry, the liver is the primary site of lipid metabolism, synthesizing over 90% of fatty acids. Very low-density lipoprotein (VLDL) plays a pivotal role in this process, transporting endogenous triglycerides (TG) from the liver to various tissues [2]. A specific subtype of VLDL, VLDLy, which includes apolipoprotein-VLDLII (ApoVLDLII) and apolipoprotein B100, is critical for moving endogenous triglycerides from the liver to the develo** chicken oocyte [3]. Excessive fat accumulation in the liver can lead to lipid metabolic disorders and fatty liver hemorrhagic syndrome in hens [4, 5]. Thus, effectively regulating lipid metabolism in aging laying hens is essential for prolonging their productive lifespan and maintaining their health [6].

It was reported that estrogen levels may be a contributing factor to the disorders of lipid metabolism in late-stage laying hens [7]. Previous studies have demonstrated that serum estrogen levels in hens decrease with age after sexual maturity [8, 9]. Natural flavonoid, particularly isoflavone which exhibit xeno-estrogenic effects, capable of binding to estrogen receptors to elicit estrogenic or anti-estrogenic responses [10,11,12,13,14]. Moreover, dietary supplementation with genistein has been shown to alleviate fatty liver conditions in hyperlipidemic laying hens through up-regulation of estrogen receptor alpha (ERα) expression while inhibiting the expression of the NLR family pyrin domain containing 3 (NLRP3) inflammasome [11]. Additionally, the inclusion of quercetin in diets has been found to boost antioxidant status and hormone levels, thereby improving production performance in late-laying hens [13]. These findings suggest that dietary flavonoids may be an effective strategy for regulating lipid metabolism in aging laying hens.

Silymarin (SIL), a flavonolignan extracted from the seeds of milk thistle (Silybum marianum L. Gaertn), is renowned for its hepatoprotective properties, antioxidant activity, immunoregulatory effects, and its ability to regulate lipid metabolism [15,16,17,18]. It can bind to estrogen receptors across various tissues, leading to diverse physiological effects [19]. Our previous studies have shown that silymarin supplementation significantly influenced lipid and bile acid metabolism in yellow feather broilers [20, 21]. Faryadi et al. [22] reported that adding lecithinized silymarin and nano-silymarin could improve the production performance of laying hens, but this research didn't articulation the specific mechanisms. Additionally, recent findings indicate that silymarin mitigated obesity in mice fed a high-fat diet by modulating gut microbiota and its metabolites [23]. Nevertheless, the impact of SIL on the intestinal microbiota of late-stage laying hens has not been documented by any researchers. Building on these findings, we hypothesize that silymarin can improve production performance and egg quality in late-stage laying hens by regulating liver lipid metabolism and the metabolites of intestinal microbiota, and demonstrated the feasibility of the experiment and discussed the experimental dosage. This study aims to examine the dose-dependent effects of silymarin supplementation and to elucidate the underlying mechanisms by which silymarin alters lipid metabolism. These insights could provide a theoretical basis for enhancements in the laying hens farming industry.

Materials and methods

Birds and experimental design

Hunan Agricultural University Animal Ethics Committee (Changsha, China) reviewed and approved all experimental protocols (HAU ACC 2022176).

A total of 480 68-week-old Lohmann Pink layers with similar body weight (2.00 ± 0.12 kg) and laying rate (92.31% ± 1.12%) were randomly assigned to 5 groups (96 birds per group) with 6 replicates per group and 16 birds each replicate. The basal diet was not supplemented with other feed additives, coccidiostats, antibiotics, only non-starch polysaccharide enzyme, non-heat-treated powdered feed, and then the basal diet was additionally mixed with different levels of silymarin. The control (CON) group were fed with basal diet, and experimental groups were fed the basal diet with 250, 500, 750 or 1,000 mg/kg silymarin powder (SIL250, SIL500, SIL750, and SIL1000), respectively. The basal diet was formulated according to GB/T 5916–2020 recommendations (Table 1). All birds was raised in cages (188 cm × 34 cm × 37 cm) with 16 birds per cage in an environmentally controlled house that maintained at 24 ± 2 °C with a relative humidity of 45% to 60%. All the hens were free access to food and water with exposure to 16 h of light/d. Silymarin used in this study was provided by Inner Mongolia Ever Brilliance Biotechnology Co., Ltd. (Inner Mongolia, China). Notably, Silymarin is a standard extract from milk thistle, and its main components include silymarin as well as several flavonolignans, which are silymarin (47.09%), silybinA + B (27.82%), isosiybinA + B (5.24%), silydianin (12.01%), silychristin (1.93%), and the detailed information of the compounds were provided in Fig. S1.

Table 1 Ingredient composition and nutrient level of the basal diets (dry basis)
Full size table

Laying performance and egg quality analysis

At the whole trial period, the egg production and weight of each replicate were recorded daily, and feed intake were measured by replicate every 4 weeks (n = 96). Average daily feed intake (ADFI) and feed to egg ration were calculated every 4 weeks (n = 96). At the end of weeks 4, 8, and 12 of the trial, 5 eggs were randomly collected from each replicate (n = 30) for determining egg quality index including egg weight, albumen height, yolk color and haugh unit by ORKA EA-01 egg quality analyser (ORKA Food Technology Ltd., Herzeliya, IL) and determining eggshell strength by EFR-01 eggshell strength tester (ORKA Food Technology Ltd., Herzeliya, IL). Yolk weight, albumen weight and eggshell weight were measured by an electronic balance (SF-400, BAIJIE, China). Eggshell thickness without inner membrane was determined at blunt end, tip end and equatorial region using a vernier caliper (ARZ-1331, AIRAJ, China) and calculated the average of thickness. The eggshell brightness (L) value was measured once for each egg on the equatorial region by a precision colorimeter (NR20XE, 3nh, China).

Serum, liver tissue and cecal content samples collection

At the end of the weeks 4, 8, and 12 of the trial, 1 bird from per replicate was randomly selected after 12 h food withdrawal to collect blood samples from the wing vein. Serum was separated from blood samples by centrifugation at 3,000 × g for 15 min using a centrifuge (TDZ4-WS, MKE, China) and then were stored at –20 °C for further analysis. The laying hens were killed by cervical dislocation and removed intact liver tissues to photograph on a clean background. After flushing the liver with saline, liver sample about 1 cm3 in size were collected from right lobe and placed in a 10% neutral formalin solution for liver morphology examination. A portion of the liver tissue was excised, immediately frozen in liquid nitrogen, and subsequently stored at –80 °C for mRNA analysis. The tip of the cecum was carefully excised from the laying hens, and the contents were expressed into sterile freezer tubes. These samples were then promptly frozen in liquid nitrogen and preserved at –80 °C, designated for subsequent metagenomic and targeted metabolomic analyses.

Serum biochemical parameters

After the serum samples (n = 6) were pre-processed, serum total cholesterol (TC), triglycerides (TG), high-density lipoprotein-cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C), alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were determined using the commercial assay kits (Nan**g Jiancheng Bioengineering Company, Jiangsu, China) according to the manufacturer’s protocols. Serum VLDL was analyzed using an ELISA kit (Shanghai Meilian Bio-technology Ltd., China) by enzyme-linked immunosorbent assay. And absorbance readings were taken for the treated samples using enzyme marker (Infinite M PLEX, Tecan, Switzerland).

RNA isolation and real-time quantitative PCR

Total hepatic RNA was extracted from liver tissue (n = 6) using TRIzol reagent (Vazyme, Jiangsu, China) following the manufacturer’s instructions. The purity and concentration of RNA was determined using NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The high-capacity cDNA Reverse Transcription kit (AG, Hunan, China) was used for cDNA synthesis. Primer sequences used in the study were present in Table 2. Real-time quantitative PCR was conducted using SYBR Green Pro Taq HS (AG, Hunan, China) using a Real-time PCR machine (LightCycler 480 II, Roche, Switzerland). The reaction conditions were as follows: 50 °C for 2 min, 95 °C for 10 min; 40 cycles of 95 °C for 15 s, 60 °C for 1 min. Each sample was measured in duplicate and the relative mRNA expression of target genes was calculated using β-actin as an internal control by the 2−ΔΔCT method. The Ct value of the target gene was obtained using RT-qPCR with β-actin as the internal gene. ΔCt = Ct (target gene) – Ct (β-actin). Normalized with the ΔCt of CON group, the variance multiplier was calculated by 2−ΔΔCT in the final step.

Table 2 Primers used for RT-qPCR
Full size table

Microbial DNA extraction and metagenomics analysis

Total microbial genomic DNA from cecal content (n = 6) was extracted using an OMEGA Soil DNA kit (D5625-01, OMEGA, USA). The quantity and quality of extracted DNA were measured using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and agarose gel electrophoresis, respectively. The extracted microbial DNA was processed to construct metagenome shotgun sequencing libraries with insert sizes of 400 bp by using Illumina TruSeq Nano DNA LT Library Preparation kit (Illumina, USA). Each library was sequenced by Illumina HiSeq X-ten platform (Illumina, USA) with PE150 strategy at Personal Biotechnology Co., Ltd. (Shanghai, China).

Raw sequencing reads were processed to obtain quality-filtered reads for further analysis. First, sequencing adapters were removed from sequencing reads using Cutadapt (v.1.2.1) [24]. Secondly, low quality reads were trimmed using a sliding-window algorithm in fastp [25]. Thirdly, reads were aligned to the host genome of broiler using BMTagger to remove host contamination. Once quality-filtered reads were obtained, taxonomical classifications of metagenomics sequencing reads from each sample were performed using Kraken2 [26] against an RefSeq-derived database, which included genomes from archaea, bacteria, viruses, fungi, protozoans, metazoans and Viridiplantae. CDS sequences of all samples were clustered by mmseqs2 [27] with “easy-cluster” mode, setting protein sequence identity threshold to 0.90 and covered residue of the shorter contig to 90%. The high-quality reads of each sample were aligned against the gene catalog by Salmon [28] to calculate relative gene abundance.

Beta diversity analysis was performed to investigate the compositional and functional variation of microbial communities across samples using Bray–Curtis distance metrics and visualized via principal coordinate analysis (PCoA). Based on the taxonomic and functional profiles of non-redundant genes, linear discriminant analysis effect size (LEfSe) was performed to detect differentially abundant taxa and functions across groups using the default parameters (P < 0.05, LDA thresholds > 2). The functionality of the non-redundant genes was obtained by annotated using MMseqs2 with the “search” mode against the protein databases of KEGG, EggNOG and CAZy, respectively, and using the Student's test method to compare the differences in abundance of each functional unit between sample groups. P < 0.05 was considered to be statistically significant, whereas a P < 0.10 was considered to constitute a tendency.

Targeted metabolomics analysis of cecal chyme

The frozen cecal contents (n = 6) were mixed with pre-cooled methanol/acetonitrile/water solution (2:2:1, v/v). The mixture was milled (60 Hz, 2 min) and then subjected to ultrasonic extraction in an ice-water bath for 10 min. The mixture was allowed to stand at –20 °C for 10 min and then centrifuged for 20 min (14,000 × g, 4 °C) and the supernatant was dried under vacuum. For mass spectrometry analysis, the sample was added with 100 μL of acetonitrile aqueous solution (acetonitrile:water = 1:1, v/v) and centrifuged for 15 min (14,000 × g, at 4 °C) and the supernatant was taken into the sample for analysis. Samples were separated on an Agilent 1290 Infinity LC UHPLC system with HILIC and C18 columns; the HILIC column temperature was 35 °C, the flow rate was 0.3 mL/min, and the injection volume was 2 μL; mobile phase composition A: 90% water + 2 mmol/L ammonium formate + 10% acetonitrile, B: acetonitrile + 0.4% formic acid. The AB 6500 + QTRAP mass spectrometer (AB SCIEX, USA) was used for mass spectrometry analysis. The ESI source conditions were as follows: source temperature: 580 °C, Ion Source Gas1 (GS1): 45, Ion Source Gas2 (GS2): 60, Curtain Gas (CUR): 35, IonSpray Voltage (IS): + 4,500 V or –4500 V in positive or negative modes, respectively, monitoring using the MRM model.

The peaks were extracted from the MRM raw data using MultiQuant or Analyst software. The ratio of the peak area of each substance to the peak area of the internal standard was then obtained, and the content was calculated according to the standard curve. Differential metabolites were identified by calculating the amount in the sample. The processed data were analyzed using the R software packages (ropls, V1.22.0; ggplot2, V3.4.1; pheatmap, V1.0.12; corrplot, V4.0.3), where they were subjected to multivariate data analysis, including principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA). Finally, the pathways in which differential metabolites were involved were obtained through metabolic pathway annotation in the KEGG database.

Statistical analysis

Statistical analysis was performed with SPSS 26.0 software (IBM, USA). Data were tested for normal distribution using the Kolmogorov–Smirnov (K-S) test. The non-normally distributed data was transformed using their respective square roots, while the normally distributed data was left unchanged. Differences between groups were examined using one-way analysis of variance (ANOVA) with Duncan’s post hoc test, when the homogeneity of variance is not significant (P > 0.05). When the homogeneity of variance is significant (P < 0.05), using Welsh ANOVA test with Tamhane post hoc test. The linear and quadratic effects of dietary SIL supplementation dose were evaluated by regression analysis. All the results were expressed as mean ± standard deviation (SD). Differences were considered significant at P < 0.05.

Results

Production performance

Table 3 demonstrated that the dietary addition of SIL did not significantly influence the overall laying rate when compared to the CON group (P > 0.05). A quadratic increase in laying rate was observed (P = 0.017) with incremental additions of SIL during the first four weeks of the experiment trial. The laying rate in the CON group significantly decreased during weeks 9 to 12 compared to weeks 1 to 4 and weeks 5 to 8 (P = 0.042). In contrast, the groups receiving SIL supplementation exhibited no significant changes in laying rates between these time periods (P > 0.05), as shown in Fig. 1. Relative to the CON group, the treatments with SIL500 and SIL750 significantly enhanced average egg weight from weeks 5 to 8 (P = 0.049). Furthermore, the average daily feed intake for hens in the SIL750 group was significantly higher throughout the trial period compared to the CON group (P < 0.05). The feed-to-egg ratio in the SIL500 group showed a significant reduction from weeks 5 to 8 (P = 0.003).

Table 3 The effects in dietary silymarin inclusion on laying performance
Full size table
Fig. 1
figure 1

Changes in laying rate over the duration of the experiment in CON and SIL groups. a,bValues with no common superscripts indicate a significant difference (P < 0.05), NS, no significant difference

Full size image

Egg quality

As shown in Table 4, shell strength increased quadratically (P = 0.025) in response to the increasing addition of SIL by week 4. However, other indices did not differ significantly among the groups at week 4 (P > 0.05). By week 8, compared with CON group, the SIL500 and SIL750 treatments significantly improved eggshell thickness (P < 0.001). Meanwhile, the eggshell brightness in the SIL750 and SIL1000 groups was lower (P = 0.013). By week 12, the SIL1000 treatment had significantly reduced the eggshell thickness, eggshell weight (P < 0.001), and eggshell brightness (P = 0.003) compared with the CON group.

Table 4 The effect of silymarin on the egg quality of laying hens
Full size table

Serum biochemistry

As shown in Table 5, at week 4, serum AST activity was significantly lower in the SIL250 and SIL750 groups compared to the CON group (P = 0.002), with a quadratic decline observed in AST activity (P = 0.005). Serum VLDL content increased linearly (P < 0.001) in response to increasing doses of SIL. By week 8, serum AST activity in the SIL500 and SIL750 groups was significantly improved compared to other groups (P < 0.001), with AST activity increasing quadratically (P = 0.007). By week 12, the SIL500 group exhibited a decrease in serum AST activity compared to other groups (P = 0.002). Additionally, with increasing doses of SIL, there was a linear decrease in both serum ALT activity (P = 0.013) and VLDL content (P < 0.001). Serum ALT activity in the SIL500 and SIL1000 groups was significantly lower than in the CON group at week 12 (P = 0.048).

Table 5 The effect of silymarin on the serum biochemistry of laying hens
Full size table

Regarding serum lipid profiles, at week 4, serum TG levels in the SIL1000 group were significantly higher than in the other groups (P = 0.034) and LDL-C content increased linearly with increasing doses of SIL (P < 0.001). At week 12, TC and TG levels decreased quadratically (P < 0.05), while HDL-C and LDL-C content decreased linearly (P < 0.001) with increasing additions of SIL. At week 12, the dietary inclusion of 500, 750, or 1,000 mg/kg of silymarin significantly reduced serum TC (P = 0.023) and LDL-C content (P < 0.001) compared to the CON group. Serum TG levels in the SIL250, SIL500, and SIL750 groups were significantly lower than in the CON group (P = 0.015). Compared to the CON group, the addition of silymarin in the diet led to a decrease in serum HDL-C concentration (P < 0.001) and an increase in serum VLDL content (P < 0.001).

Liver lipid metabolism

As illustrated in Fig. 2, H&E staining of liver sections from the CON group revealed lipid vacuolization and severe steatosis in hepatocytes (Fig. 2A). Silymarin treatment effectively ameliorated liver steatosis and reduced lipid vacuolization within hepatocytes.

Fig. 2
figure 2

The effect of dietary silymarin on hepatic histomorphology of laying hens. A The effect of dietary silymarin on hepatic histomorphology H&E stained sections of laying hens (400 × magnification). B Liver images at week 12

Full size image

The impact of silymarin on the expression of lipid synthesis-related genes is depicted in Fig. 3A and C. The expression of fatty acid synthase (FASN) was regulated quadratically throughout all trial periods (P < 0.001), with significant downregulation observed in the SIL1000 group at week 4 (P < 0.001). By week 12, FASN expression in all SIL groups was significantly upregulated compared to the CON group (P < 0.001). The mRNA expression of stearoyl-CoA desaturase (SCD) also increased quadratically during all trial periods (P < 0.001), with significant increases in the SIL250 and SIL500 groups at week 4 and in the SIL500 and SIL750 groups at week 12 compared to the CON group (P < 0.001). Supplementation with 250, 500, or 750 mg/kg silymarin significantly improved the expression of elongation of very long-chain fatty acids protein 6 (ELOVL6) and inhibited ELOVL7 expression at week 4. However, ELOVL7 expression in the SIL500 and SIL750 groups was significantly lower than in the CON group at week 12 (P < 0.001).

Fig. 3
figure 3

The effect of silymarin on liver lipid metabolism of late laying hens. A and B The effect of silymarin on liver lipid metabolism of laying hens at the fourth week. C and D The effect of silymarin on liver lipid metabolism of laying hens at the twelfth week. a–cvalues with no common superscripts indicate a significant difference (P < 0.05). FASN Fatty acid synthase, SCD Stearoyl-CoA desaturase, ELOVL6 Elongation of very long chain fatty acids 6, ELOVL7 Elongation of very long chain fatty acids 7, ACC Acetyl-CoA carboxylase, PPAR-α Peroxisome proliferators-activated receptor-α, PPAR-γ Peroxisome proliferators-activated receptor-γ, SREBP-1 Sterol-regulatory element binding protein-1

Full size image

As shown in Fig. 3B and D, mRNA expression of acetyl-CoA carboxylase (ACC) increased linearly in all trial periods (P < 0.001), with the SIL500 and SIL750 groups exhibiting significantly higher levels than the CON group at both weeks 4 and 12 (P < 0.001). Dietary silymarin addition significantly decreased the expression of peroxisome proliferator-activated receptor-α (PPAR-α) at week 4, but increased its expression by week 12. The expression of sterol-regulatory element-binding protein-1 (SREBP-1) was linearly regulated throughout all trial periods (P < 0.001), with the SIL250 group showing significantly higher expression, whereas the SIL500, SIL750, and SIL1000 groups had lower expression than the CON group at week 4 (P < 0.001). Moreover, SREBP-1 expression in the SIL500 group was significantly higher than in the CON, SIL250, and SIL1000 groups at week 12 (P = 0.04).

Lipoprotein expression

At weeks 4 and 12, the apolipoprotein B (ApoB) expression regulated quadratically (P < 0.05) and SIL500, SIL750 groups were significantly higher than CON group and SIL250 group (P < 0.001) in all trial periods. ApoVLDLII and vitellogenesis (VTG) expression in all SIL groups were significantly higher than CON group at week 4 (P < 0.01). Compared with CON group, the expression of G protein-coupled estrogen receptor 30 (GPR30) regulated linearly in all trial periods (P < 0.001) and SIL250, SIL500 and SIL1000 groups were significantly reduced at week 4 (P < 0.001). ApoVLDLII and GPR30 expression in SIL500 group were significantly higher than other groups at week 12 (P < 0.01). VTG expression in SIL750 group were significantly higher than other groups at week 12 (P < 0.001) (Fig. 4A and B).

Fig. 4
figure 4

The effect of dietary silymarin on liver lipoprotein synthesis of laying hens. A The effect of silymarin on liver Lipoprotein synthesis of laying hens at the fourth week. B The effect of silymarin on liver Lipoprotein synthesis of laying hens at the twelfth week. a–cValues with no common superscripts indicate a significant difference (P < 0.05). ApoB Apolipoprotein B, ApoVLDLII Apolipoprotein Very Low Density LipoproteinII, VTG Vitellogenesis, GPR30 G protein-coupled estrogen receptor 30

Full size image

Liver bile acids metabolism and estrogen receptors expression

The mRNA expression of cholesterol 7 alpha-monooxygenase (CYP7A1) increased quadratically in all trial periods (P < 0.001) and in SIL250 and SIL1000 groups were significantly lower than other groups at weeks 4 and 12 (P < 0.001). The farnesoid X receptor (FXR) expression in SIL500 and SIL1000 were lower than other groups at week 4 (P < 0.001). The expression of FXR levels increased quadratically (P < 0.001), SIL500, SIL750 and SIL1000 groups were significantly higher than CON and SIL250 groups at week 12 (P < 0.001). The multidrug resistance associated-protein (MRP2) expression in SIL250, SIL500 and SIL1000 were significantly lower than the CON group at week 4 (P < 0.001), but at week 12 its expression in SIL750 were significantly higher than other groups (P < 0.001). Compared with CON group and SIL250 group, the bile salt export pump (BSEP) expression increased linearly in all trial periods (P < 0.001) and SIL500, SIL750 and SIL1000 groups were significantly increased (P < 0.001) at weeks 4 and 12 (Fig. 5A and B).

Fig. 5
figure 5

The effect of dietary silymarin on liver BAs metabolism and ER expression in laying hens. A The effect of silymarin on liver bile acid metabolism of laying hens at the fourth week. B The effect of silymarin on liver bile acid metabolism of laying hens at the twelfth week. C The effect of dietary silymarin on liver ERα expression in laying hens. D The effect of dietary silymarin on liver ERβ expression in laying hens. a–dValues with no common superscripts indicate a significant difference (P < 0.05). ERα Estrogen receptor α, ERβ Estrogen receptor β, CYP7A1 Cholesterol 7alpha-monooxygenase, FXR Farnesoid X receptor, MRP2 Multidrug resistance associated-protein, BSEP Bile salt export pump

Full size image

Compared with the CON group, SIL treatment significantly reduced estrogen receptor α (ERα) expression at the weeks 4 and 12 (P < 0.001) (Fig. 5C and D). The mRNA expression of estrogen receptor β (ERβ) in SIL250, SIL500 and SIL100 groups were significantly downregulated compared to CON group at week 4 (P < 0.001). At week 12, the expression of ERβ in all SIL groups were significantly improved relative with CON group (Fig. 5C and D) (P < 0.001).

Microbial composition and function

As shown in Fig. 6A, alpha diversity analysis including Chao1, ACE, Shannon and Simpson were not significant between CON and SIL groups (P > 0.05). PCoA showed that there was no obvious separation between CON and SIL groups (Fig. 6C). Bacteroidota, Firmicutes and Actinobacteria were ranked as the top 3 bacteria at the phylum level(Fig. 6B). LefSe analysis showed that Phocaeicola_gallinaceus, Phocaeicola_pullistercoris, Limosilactobacillus_sp012843675, Pelethomonas_sp017887695, Coprousia_avicola and Flavonifractor_avistercoris were regarded as the dominant bacteria in SIL group (Fig. 6D). Functional difference analysis showed four pathways that tended to be significant different between CON group and SIL group (P < 0.10), including secondary bile acid biosynthesis (ko00121), fructose and mannose metabolism (ko00051), ribosome (ko00030) and Pentose phosphate pathway (ko00030) (Fig. 6F). We further analyzed the expression of microbial genes involved in secondary bile acid metabolism pathway and found that choloylglycine hydrolase and 7-alpha-hydroxysteroid dehydrogenase were down-regulated (Fig. 6G).

Fig. 6
figure 6

Diversity analysis, differential and functional enrichment analysis of cecal microbes. A The alpha diversity of cecal microbiota was analyzed by Chao1, Simpson, Shannon, ACE. Red is CON group and blue is the SIL group. B Differences of cecal flora at phylum level. C The beta diversity of cecal microbiota was analyzed by PCoA. D The different microbial groups in cecum of laying hens were analyzed based on LEfSe method. E The circos map of functional component relationships F CON group vs. SIL group comparison between different functional units. G The comparison of gene expression enriched in secondary bile acid biosynthesis (pink for CON group, yellow for SIL group). PCoA, Principal coordinates analysis; LEfSe, Linear discriminant analysis effect size

Full size image

Metabolomic profiling of the cecum

As shown in Fig. 7, the PLS-DA score plot and OPLS-DA score plot both showed a clear separation of metabolic profiles between the CON and SIL groups (Fig. 7A and B). The metabolomic analysis identified a total of 366 metabolites (Fig. 7C). In SIL group, we identified 9 differential metabolites compared with CON group (5 in positive ion mode and 4 in negative ion mode; Fig. 7C). The content of ricinoleic acid, thiamine monophosphate, glucosamine-6-phosphate, hydroxyproline and cis-4-Hydroxy-D-proline were significantly higher in SIL group than CON group, but uridine 5′-diphosphate (UDP), N2,N2-dimethylguanosine, vitamin B7 and ketoleucine concentrations were lower in SIL group than CON group (Fig. 7D). A further KEGG enrichment analysis showed that the different metabolites were closely related to valine, leucine and isoleucine biosynthesis, biotin metabolism, alanine, aspartate and glutamate metabolism, thiamine metabolism, arginine proline metabolism and the ATP-binding cassette (ABC) transporters pathway. Arginine proline metabolism and ABC transporters were the most significant (Fig. 7E).

Fig. 7
figure 7

Metabolic alterations induced by silymarin. A PLS-DA between the SIL and CON groups. B OPLS-DA score plot between the SIL and CON groups. C Volcano plot shows the different metabolites between the SIL and CON groups. D Heat map shows the alteration patterns of significantly changed metabolites between the SIL and CON groups. E KEGG pathway analysis enriched form all differential metabolites. F Correlation heatmap between top 20 abundance of bacteria and top 20 differential metabolites (P < 0.068). PLS-DA, Partial least squares discrimination analysis; OPLS-DA, Orthogonal partial least squares-discriminant analysis

Full size image

Discussion

Enhancing the production performance and egg quality of late-stage laying hens is crucial for increasing their economic value [29]. The liver, the crucial organ in lipid metabolism [5], is particularly susceptible to lipid accumulation and oxidative stress during this period [1]. These conditions can adversely affect the production performance of the hens. Previous studies have indicated that dietary supplementation with silymarin not only improves the laying rate and egg weight but also reduces the feed-to-egg ratio [23]. Similar improvements in production performance have been observed in broilers when silymarin is added to their diets [20, 30]. Recent research supports these findings, showing that 500 mg/kg of dietary silymarin significantly enhances average egg weight and lowers the feed-to-egg ratio, aligning with earlier reports. Eggshell strength and thickness are critical indicators of egg quality. Nie et al. [31] demonstrated that eggshell quality is linked to the absorption of calcium ions in the small intestine of laying hens. In this study, the addition of 500 mg/kg of silymarin appeared to enhance eggshell thickness and strength, likely by improving intestinal absorption capabilities [32].

Interestingly, we observed a significant decrease in the laying rate of the CON group as the experiment progressed. In contrast, the laying rates in the SIL groups remained stable, suggesting that silymarin mitigates the decline in laying rates among late-stage laying hens. The liver, crucial for lipid metabolism and yolk precursor production [33], often exhibits health through the levels of transaminases such as ALT and AST in hepatocytes [34]. This study found that a dietary addition of 500 mg/kg silymarin reduced serum ALT and AST levels by week 12. Consistent with our findings, previous studies reported reductions in serum ALT and AST levels in broilers exposed to mycotoxins [16] and in models of chemically induced liver damage [35].

Excessive fat accumulation in hepatocytes can lead to histological changes in the liver. Serum TG and TC levels were important indexes to explain the capacity of lipid metabolism in animals. In our experiment, SIL treatment significantly lowered serum TG and TC level. These results align with findings by Ghazaghi et al. [36], who noted a significant decrease in serum TC in quails treated with silymarin, and a recent study showing reduced serum TC and LDL-C in broilers [20]. Moreover, HDL and LDL play essential roles in hepatic lipid homeostasis [37]. Similar to previous studies [38, 39], our research found that silymarin supplementation decreased serum LDL-C and HDL-C levels in late laying hens by week 12. Specifically, we observed an increase in serum VLDL levels, which are critical for transporting endogenous triglycerides from the liver to tissues [40]. Further investigation into liver expressions of ApoB, ApoVLDLII, VTG, and GPR30 which are involved in VLDL assembly and yolk precursor transport [41, 42] revealed that silymarin enhances VLDL assembly and yolk precursor synthesis. These findings suggest that silymarin could improve the production performance of late-stage laying hens by alleviating liver steatosis and enhancing the synthesis of lipoproteins and yolk precursors. In this context, a 500 mg/kg dose of SIL was the most effective in regulating lipid metabolism in these hens.

To further elucidate the underlying mechanisms of silymarin on lipid metabolism in laying hens, we analyzed the expression of genes related to lipid metabolism, including ACC, FASN, PPAR-α, and SREBP-1 [38]. The ELOVL family, particularly ELOVL6 and ELOVL7, are known to respond to lipid deposition in broilers [43]. Our findings show that silymarin enhances the expression of these genes, suggesting a beneficial role in the elongation of very long chain fatty acids. Consistent with our study, previous research has indicated that silymarin and silybin can suppress genes involved in lipid synthesis, such as FASN and SCD [44]. Furthermore, silymarin has been shown to increase the expression of lipolysis-related genes such as ACC [56]. MRP2 is the major bile acid transport protein in liver, which may promote bile acid transport to the intestine [57]. The previous study indicated that silymarin plays a vital role in biliary excretion of MRP2 [58]. Silymarin also could prevent cholestasis-associated recovery of the BSEP [59]. Similar with above results, our results showed that silymarin upregulated FXR expression and downregulated the CYP7A1 expression. In our study, silymarin reduced the endogenous bile acid synthesis and accelerate the enterohepatic circulation.

Cecal microbiota has been shown to correlate with laying performance of layers [60, 61]. Lucke et al. [62] reported that the most dominant bacteria at the phylum level were Bacteroidota and Firmicutes in poultry which was consistent with our results. Silymarin inhibited the growth of pathogenic bacteria and promoted the colonization of beneficial bacteria in the intestine of poultry [63, 64]. Lipid metabolism of poultry was related to gut microbiota [65, 66]. The LefSe analysis revealed that the dominant bacteria in the SIL group were primarily associated with the lipid metabolism of the host, including s_Phocaeicola s_Limosilactobacillus_sp012843675, s_Pelethomonas_sp017887695, and s_Flavonifractor_avistercoris. Zhou et al. [67] reported that Limosilactobacillus-reuteri improved lipid metabolism via regulating gut microbiota in mice fed with a High-Fat diet. Previous studies reported that dietary supplementation with Limosilactobacillus improved performance production in poultry [68, 69]. Cristina et al. [70] found that NAFLD decreased the abundance of Flavonifractor in human intestine. Mikami et al. [71] found that oral Flavonifractor could attenuate inflammatory responses in obese adipose tissue. In our study, SIL group increase the abundance of Flavonifractor, which had a potential role in promoting lipid metabolism The above experiments demonstrated that the regulation of lipid metabolism by silymarin in late laying hens was inextricably linked to its modulation of intestinal microbial composition.

Cecal microbiota has been reported to participate in bile acid metabolism. Previous study showed that silymarin could regulate bile acid metabolism via FXR [72]. In our study, silymarin improved the relative abundance of Phocaeicola genus, which was positively correlated with unconjugated chenodeoxycholic acid and performed bile acid deconjugation [73, 74]. Ye et al. [74] reported that Lactobacillus regulated bile acid metabolism via regulating FXR signaling-mediated. Furthermore, the functional unit analysis revealed that silymarin increased the expression of microbial gene related to secondary bile acid metabolism. The results demonstrated that silymarin alleviated bile acid stagnation in the body by modulating gut microbial function and enhancing the ability of metabolic microorganisms to metabolize bile acids.

Through the metabolomics analysis, we found 5 significantly up-regulating metabolites in SIL treatment, including ricinoleic acid, thiamine monophosphate, Hydroxyproline, cis-4-hydroxy-D-proline and glucosamine 6-phosphate. Biotin could act as a coenzyme for carboxylases regulating lipid and amino acid metabolism [75]. Oloyo et al. [76] reported that dietary addition of biotin reduced serum TG, TC contents in chicks. The results of this study showed that differential metabolites biotin metabolism, which implied that biotin may contribute to the regulation of lipid metabolism by silymarin. Recent study showed that high-dose thiamine could prevent the development of experimental fatty liver driven by overnutrition [77]. Hamano et al. [78] reported that dietary thiamine supplementation reduced plasma TG in broiler chicks.

Our results showed that TMP content was significantly higher in SIL group. Therefore, it can be inferred that silymarin may enhance the absorption of thiamine by the host. In bacteria, ABC importers were involved in the uptake of nutrients and micro-nutrients through medium- and high-affinity pathways [79], which promoted cellular nutrient uptake of microorganisms and host. The results of differential metabolite KEGG pathway enrichment analysis showed significant upregulation of ABC transporters in the SIL group. These were considered to elucidate the effect of silymarin on lipid and bile acid metabolism by regulating intestinal microorganisms and their metabolites.

Conclusions

In summary, silymarin supplementation could improve performance by regulating liver lipid metabolism, estrogen receptors activity and improving cecal microbiota and its function in late laying hens. Specifically, supplementation with silymarin decreased serum TG, TC, HDL-C, LDL-C content, and increased serum VLDL content by regulating the expression of FASN, ACC, SCD, PPAR-α and ApoVLDLII in liver. Moreover, dietary silymarin regulated the expression of FXR, CYP7A1, BSEP and MRP2 in liver and altered the cecal microbiological structure and three species of Phocaeicola were dominated microbial functions which were enriched in secondary bile acid synthesis. The results highlight that silymarin is a feasible feed additive for late laying hens, with the optimal dosage being 500 mg/kg. The present study provided a theoretical basis for the application of silymarin extract in late laying hens.