All procedures in this experiment were approved by the Animal Care and Use Committee of the Ningbo Academy of Agricultural Sciences (approval number NKYLL-2024-02). Twenty healthy Large White × Yorkshire sows, aged 9 months, were randomly selected and assigned to four groups following the principle of similar weight. The experiment spanned one month. The four study groups were as follows: (1) sows fed with a basal diet (Control); (2) sows fed with a basal diet supplemented with HH [5 × 10¹⁰ colony-forming units (CFU) per sow in sterile PBS]; (3) sows fed with a basal diet supplemented with 5 × 10¹¹ CFU HH; (4) sows fed with a basal diet supplemented with 5 × 10¹² CFU HH. The concentrations of HH used in this study were based on previous studies (Reagan-Shaw et al., 2008; Li et al., 2020). Hungatella hathewayi (HZBMC, HZbio, Microbial Conservation; catalog number HZB432244) was obtained and cultured anaerobically in PYG medium (104b, Leibniz-Institut DSMZ GmbH, Braunschweig, Germany) at 37°C for 12–14 h. All sows had ad libitum access to feed and water. The nutritional composition of the sows’ diet is detailed in Table 1.

Following a one-month trial, blood samples were collected and allowed to stand at room temperature for 30 min to facilitate clotting. Subsequently, the serum was separated by centrifugation at 1,500 × g for 5 min at 4°C. Fecal samples were also collected from the sows after the trial. All samples were stored at −80 °C for subsequent analysis.

Serum samples were utilized to determine the concentrations of progestogen, follicle-stimulating hormone (FSH), and estrogen, as well as the levels of total antioxidant capacity (T-AOC), glutathione (GSH), and malondialdehyde (MDA). These analyses were performed using commercial assay kits provided by Shanghai Enzyme-linked Biotechnology Co., Ltd. Absorbance measurements of the serum samples were conducted using a microplate reader (Tecan, Männedorf, Switzerland).

The genomic DNA of sows’ feces was extracted using the MolPure^® Stool DNA Kit (YEASEN Biotech, Shanghai, China) following the manufacturer’s guidelines, extracted DNA was quantified using a NanoDrop2000 spectrophotometer (Thermo Fisher Scientific, DE, USA) and qualified using 1% agarose gel electrophoresis.

The V4 hypervariable regions of the microbiota 16S rRNA gene were amplified using specific primers 515F (GTGCCAGCMGCCGCGGTAA) and 806R (GGACTACHVGGGTWTCTAAT). All PCR reactions were conducted using 15 μL of Phusion High-Fidelity PCR Master Mix from New England Biolabs, with 0.2 μM of both forward and reverse primers and approximately 10 ng of template DNA. The thermal cycling protocol commenced with an initial denaturation step at 98°C for 1 min, succeeded by 30 cycles consisting of denaturation at 98°C for 10 s, annealing at 50°C for 30 s, and extension at 72°C for 30 s. A final extension step was performed at 72°C for 5 min. PCR products were purified using a Qiagen Gel Extraction Kit manufactured in Germany. Sequencing libraries were prepared employing the TruSeq DNA PCR-Free Sample Preparation Kit from Illumina, USA, adhering to the supplier’s guidelines, with index codes appended. The quality of the libraries was evaluated using a Qubit 2.0 Fluorometer from Thermo Scientific and an Agilent Bioanalyzer 2100 system. Ultimately, the libraries were sequenced on an Illumina NovaSeq platform, yielding 250-bp paired-end reads.

In this study, the processing of primer sequences was conducted using Vsearch (version 2.14.1) and Usearch (version 11.0.667), which are widely recognized tools for handling high-throughput sequencing data. The raw sequencing data underwent a series of steps, including trimming to remove primer sequences, filtering to eliminate low-quality sequences, dereplication to remove duplicate sequences, and denoising to correct sequencing errors, resulting in high-quality clean amplicons. These clean amplicons were then used for taxonomic annotation against the SILVA database (release 132_99), a comprehensive and authoritative database for microbial classification.

The diversity of microbial communities was evaluated through the computation of alpha and beta diversities using Usearch. Alpha diversity, which reflects the richness and evenness of microbial communities, was assessed using Shannon and richness indices. Beta diversity, which indicates the variation among different microbial communities, was analyzed using constrained principal coordinates analysis (cPcoA) to compare intestinal microbiota across different experimental groups.

To identify differentially abundant microbes between groups, STAMP (version 2.1.3) software was employed for statistical analysis. Additionally, linear discriminant analysis (LDA) coupled with the LEfSe tool was used to detect microbial biomarkers that significantly differed across groups. The LEfSe tool, accessible at https://www.bic.ac.cn/BIC/#/analysis?page=b%27MzY%3D%27&tool_type=tool (accessed on 22 October 2024), was utilized for this purpose. Finally, the phylogenetic investigation of Communities Reconstruction of Unobserved States (PICRUSt2) was employed to predict the functional potential of the fecal microbiota based on 16S rRNA gene sequences (Douglas et al., 2020).

Serum samples (100 μL) were aliquoted into EP tubes and resuspended with prechilled 80% methanol using vigorous vortexing. Following this, the samples were placed on ice for 5 min to allow for complete precipitation, then centrifuged at 15,000 g at 4°C for 20 min. A portion of the supernatant was diluted to a final concentration containing 53% methanol using LC-MS grade water. The samples were transferred to new Eppendorf tubes and centrifuged again at 15,000 g at 4°C for 20 min. The supernatant was then carefully injected into the LC-MS/MS system for positive and negative ion modes analysis to facilitate relative quantification. Metabolites were identified and quantified by comparing their retention times and mass-to-charge ratios (m/z) with those in reference databases. The metabolomic data were processed by subtracting background ions as determined by blank samples and normalizing the data. The identified metabolites were annotated using the KEGG,¹ HMDB,² and LIPIDMaps³ databases. Metabolites were deemed differential if they exhibited a Variable Importance in Projection (VIP) score exceeding 1, a P-value below 0.05, and a fold change of at least 2 or no more than 0.5. The roles of these metabolites were then further elucidated using the MetaboAnalyst online platform (version 6.0).

The hormone data of sows were processed and analyzed using one-way analysis of variance (ANOVA) with the SPSS 23.0 software package (IBM, Armonk, NY, USA). The results were visualized as the mean ± standard deviation (SD) using Prism (v.8.0; GraphPad Software, San Diego, CA, USA). Statistical significance was determined using a P-value threshold of less than 0.05.

As depicted in Figure 1, dietary supplementation with Hungatella hathewayi did not significantly alter the concentration of FSH. However, it prominently decreased the progesterone level at a dosage of 5 × 10¹¹ CFU and increased the concentration of estrogen at 5 × 10¹² CFU (P-value < 0.05). The levels of T-AOC, SOD and GSH indicated that supplementation with Hungatella hathewayi did not significantly change the host’s antioxidative capacity. Subsequently, serum and fecal samples from both the control and treatment groups that received a dose of 5 × 10¹² CFU/sow of HH were subjected to further analysis.

16S rRNA sequencing indicated that the microbial richness index increased, while the Shannon diversity index decreased in the HH group (P-value > 0.05) (Figures 2A, B). Microbial beta diversity analysis suggested alterations in fecal microbial composition (Figure 2C). At the phylum level, Firmicutes, Bacteroidetes, Spirochaetes, and Proteobacteria were the dominant phyla in both the control and HH groups (Figure 2D). At the genus level, Prevotella, Streptococcus, Clostridium_sensu_stricto, and Oscillospira were highly abundant in both groups (Figure 2E).

Linear discriminant analysis effect size (LEfSe) analysis identified the family Ruminococcaceae as a biomarker in the control group, and the family Bacteroidaceae was recognized as a key bacterium in the HH group (Figure 2F). A total of 29 genera differed significantly between the control and HH groups, with the relative abundances of Roseburia, Alloprevotella, Lachnospira, Anaerovibrio, and Hungatella being significantly increased following dietary supplementation with Hungatella hathewayi (Figure 2G). Additionally, 23 KEGG pathways, including those involved in methane and starch and sucrose metabolism, were significantly altered between groups (Figure 2H).

In our metabolomic sequencing results, a total of 3,354 metabolites were identified. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) indicated significant differentiation in the fecal metabolite profiles between the control and HH groups (Figures 3A, B). A total of 729 differentially abundant metabolites (DAMs) were identified through metabolomics analysis, 330 DAMs were significantly upregulated in the HH group, while 399 DAMs were downregulated in the control group (Figures 3C, D). KEGG pathway analysis revealed that these differential metabolites were associated with several key metabolic pathways, including pyrimidine metabolism, tryptophan metabolism, and riboflavin metabolism (Figure 3E). Furthermore, Gene Set Enrichment Analysis (GSEA) suggested that the steroid hormone biosynthesis pathway was altered between the control and HH groups (Figure 3F).

The PLS-DA model demonstrated that serum metabolites in the control and HH groups had distinct profiles in both positive and negative ion modes (Figures 4A, B). Comparative analysis revealed 183 upregulated and 70 downregulated differentially accumulated metabolites (DAMs) in the HH group relative to the control group (Figures 4C, D). These DAMs were associated with metabolic pathways such as pyrimidine metabolism, tryptophan metabolism, histidine metabolism, and beta-Alanine metabolism (Figure 4E). Furthermore, GSEA analysis indicated that the steroid hormone biosynthesis pathway was significantly altered in serum between the control and HH groups (Figure 4F).