3.1 Growth performance
Dietary supplementation with L. lactis ZLL028 significantly increased the ELR of hens by 7.29% compared with the CG. However, the FER, ADFI, and AEW did not differ significantly between the groups (P > 0.05) (Fig. 1).
3.2 Antioxidant indices and mRNA expression of CYP450 and AhR in the liver
Liver MDA levels were significantly lower, while T-AOC was elevated in the L. lactis- treated hens (LG) relative to CG. AhR (P = 0.001) and CYP2D6 (P = 0.001) mRNA levels were higher in the LG, whereas CYP1A2 expression showed no marked difference (P = 0.056). Moreover, mRNA levels of CPT-1 (P = 0.001) was markedly raised in the LG (Fig. 2).
3.3 Yolk LCFAs composition
The LCFAs profiles in the yolk, including polyunsaturated fatty acids (PUFAs), monounsaturated fatty acids (MUFAs), and saturated fatty acids (SFAs), were determined. The total MUFAs (P = 0.013) and C22:1n9 (P = 0.009) were msrkedly decreased, while the total PUFAs (P = 0.018), and C18:3n3 (P = 0.015), C18:2n6c (P = 0.020) were increased in the LG relative to the CG (Fig. 3).
3.4 Trp metabolites analysis
The LG showed markedly lower concentrations of IAA (505.78 μg/kg vs. 651.70 μg/kg, P = 0.005) and skatole (905.80 μg/kg vs. 1514.50 μg/kg, P = 0.002) but higher levels of IAld (330.71 μg/kg vs. 235.69 μg/kg, P = 0.001) compared with the CG (Fig. 4A).
In yolk samples, IAA levels were significantly decreased in the LG in comparison with the CG (17.61 μg/kg vs. 75.20 μg/kg, P = 0.002), whereas IAld concentrations were significantly higher (1099.47 μg/kg vs. 983.93 μg/kg, P = 0.004). Skatole was not detected in either group (Fig. 4B).
3.5 The GM community affected by L. lactis
The GM composition of the feces at day 0 and 30, and GM composition of ileum and cecum at day 60 were analyzed at the end of the experiment. Similarities and differences in GM composition between treatment groups were assessed at the operational taxonomic unit (OTU) level using PLS-DA (Fig. 5A-C). The LG was significantly different from the CG in fecal samples at day 30, as well as in ileum and cecum samples at the end of the experiment. The Venn diagram indicated that 1347 OTUs were common to all samples, while the LG exhibited a higher number of unique OTUs in fecal samples at day 30 (1894 vs. 344), as well as in ileum (36 vs. 7) and cecum (155 vs. 116) samples at the end of the experiment (Fig. 5D). Alpha diversity, measured using the Chao index (Fig. 3B), was higher in fecal samples at day 30 and in ileum samples in the LG compared with the CG (Fig. 5E).
At both day 30 and day 60, Bacillota, Bacteroidota, Actinomycetota, and Pseudomonadota were the abundant phyla in fecal and cecal samples, whereas Pseudomonadota, Actinomycetota, Chlamydiota, and Chordata were predominant in ileum samples. At day 30, the Bacillota abundance was significantly lower in the LG in comparison with the CG (51.64% vs. 59.59%, P = 0.008), whereas Actinomycetota abundance was higher (18.49% vs. 8.29%, P = 0.028) in fecal samples. At the end of the experiment, Chordata abundance in ileum samples was significantly lower in the LG (58.79% vs. 66.63%, P = 0.047), whereas Actinomycetota abundance was higher (5.43% vs. 0.18%, P = 0.033). In cecal samples, Bacillota abundance was significantly higher (30.50% vs. 27.56%, P = 0.008), while Bacteroidota abundance was lower (55.59% vs. 58.45%, P = 0.028) in the LG compared with the CG. At the genus level, in fecal samples at day 30, the abundance of Alistipes (0.59% vs. 0.46%, P = 0.038) and Acinetobacter (0.35% vs. 0.08%, P = 0.018) was higher, whereas Escherichia abundance was lower (1.71% vs. 2.63%, P = 0.013) in the LG in comparison with the CG. At the end of the experiment, Lactobacillus (0.57% vs. 0.31%, P = 0.037) and Alistipes (4.62% vs. 3.92%, P = 0.024) were markedly more abundant in ileum samples in the LG (Fig. 6A-D).
3.6 Metabolomics analysis
Metabolomic analysis was performed using a significance threshold of P < 0.05, an FC >1.2 or <0.8, and a VIP score >1.0. A total of 809 metabolites, including 475 positive ions and 334 negative ions, were detected. The KEGG database was used to identify pathways associated with the detected metabolites. The enriched compounds were primarily classified as phospholipids, amino acids, and other metabolites (Fig. 7A). KEGG pathway enrichment analysis indicated significant involvement in lipid metabolism, amino acid metabolism, and digestive system, among other pathways (Fig. 7B). Further classification using the HMDB revealed the presence of 11 indoles and derivatives, accounting for 1.74% of the total metabolites (Fig. 7C). Identified positive ion metabolites included indoline, indole-3-carboxaldehyde, 3-methylindole, indole, isoindoline, IAld, serotonin, indole-3-acetic acid, 3-indoleacrylic acid, and indole-3-carboxaldehyde, while quinolinic acid was identified as a negative ion metabolite (Supplementary Table 1).
3.7 Hepatocytes proliferation, antioxidant analysis in vitro by L. lactis and IAld
Hepatocyte viability and proliferation were significantly increased following treatment with L. lactis, but no marked variations were found between L. lactis and IAld in proliferation rate (147.1% vs. 149.1%) (Fig. 8A). Furthermore, MDA levels in hepatocytes significantly decreased (P = 0.020), while T-AOC significantly increased (P = 0.020) in the LG and IAld group (IG) in comparison with the CG. However, the T-AOC index was higher in the IG compared with the LG (5.71 U/mL vs. 4.40 U/mL), whereas the MDA level was seen lower in the IG than the LG group (1.01 nmol/mL vs. 0.81 nmol/mL) (Fig. 8B).
3.7.2 RNA-seq analysis
DEGs were defined as having a threshold of P ≤ 0.05 and absolute |logFC| ≥ 1. Gene expression comparisons revealed a total of 30, 108 detected genes across the CG, LG, and IG. Overall, 810 DEGs were detected in the LG vs. CG, including 491 upregulated and 319 downregulated genes. In the IG vs. CG, 281 DEGs were detected, with 66 upregulated and 215 downregulated genes. The IG and LG comparison identified 1, 994 DEGs, of which 628 were upregulated and 1, 366 downregulated.
GO annotation classified DEGs into the categories of cellular component, molecular function, and biological process. The LG vs. CG had 19, 2, and 19 GO terms in these categories, respectively, while the IG vs. CG contained 19, 2, and 20 GO terms. The IG vs. LG had 17, 2, and 10 GO terms. GO clustering analysis revealed that GO terms ‘cellular process’, ‘binding’, and ‘cellular anatomical entity’ were significantly upregulated in both the LG and IG compared with the CG (Fig. 9A-C).
Pathway analysis was performed based on the KEGG pathway database to predict the significantly enriched metabolic. It identified 304, 159, and 335 enriched KEGG pathways in the LG vs. CG, IG vs. CG, and IG vs. LG, respectively. Signal transduction pathways, particularly the signal transduction, and signaling molecules and interaction were markedly enriched in the LG vs. CG, IG vs. CG, and IG vs. LG (Fig. 9D-F).
3.7.3 Cytokine detection of p38/NF-κB axis in hepatocytes by L. lactis and IAld
Then, we analyzed the expression of key genes in the p38, NF-κB and Nrf2 signal pathway, and the results showed that mRNA levels of p38, NF-κB were significantly decreased and Nrf2 was significantly increased in hepatocytes in the LG and IG compared to the CG (P = 0.001). No significant variation was seen between IG vs. LG regarding p38 mRNA expression (Fig. 10 A-C).