Campylobacter and Acinetobacter, known to harbor the HEMESYN2-PWY gene cluster (hemA-hemH), exhibited significantly correlations with the HEMESYN2-PWY (heme biosynthesis II), which was enriched in the SO stage (postpartum sows) (Fig. 6). Furthermore, Campylobacter, Acinetobacter, Psychrobacter, Chryseobacterium, and Pseudomonas were highly correlated with the PWY-3781 (aerobic respiration I), REDCITCYC (TCA cycle VIII), and PWY0-1061 (superpathway of L-alanine biosynthesis), all of which were enriched in the O4 stage (Figs. 6 and 7a). Through interpathway correlation analysis based on 19 KEGG functional pathways from PICRUSt and LEfSe functional analyses, we identified a strong association between the HEMESYN2-PWY/heme biosynthesis II (anaerobic) which was enriched in the S0 (postpartum sows) stage and the PWY-3781 (aerobic respiration I), REDCITCYC (TCA cycle VIII), and PWY0-1061 (superpathway of L-alanine biosynthesis), all of which were enriched in the O4 stage. Additionally, pathways enriched in the O1 stage showed strong correlations with other pathways within the same O1 stage. (Fig. 7b).
In our study, sequencing data provided valuable insights into how maternal sources and age influence the early microbial establishment and dynamics of the pig gut microbiota. Early gut microbiota colonizers are essential for establishing a mature microbial community, ultimately influencing the health and productivity of pigs. Despite growing research on pig gut microbiota, few studies have examined the early development of piglet gut microbiota and even fewer have systematically identified maternal effects during early life. To address this gap, we conducted a large scale study investigating the influence of maternal sources on the early microbial establishment of the piglet gut microbiota from birth. Additionally, we analyzed the dynamic changes in microbiota composition, network interactions, and metabolic functions across piglet development, including the postpartum stage of the sow. Early in life, during the lactation stage (O1, Day 5), Escherichia-Shigella, Bacteroides, and Fusobacterium were core microbiota in piglet gut microbiota. Our findings highlighted the critical impact of maternal fecal microbiota on piglet gut composition, which was associated with a higher transfer rate of Escherichia-Shigella, Bacteroides, and Fusobacterium. Specifically, Bacteroides is linked to the utilization of oligosaccharides in milk. Surprisingly, this study revealed that not only Bacteroides, a key genus involved in breast milk digestion, but also Escherichia-Shigella and Fusobacterium, previously believed to originate from the environment, actually originate from the sow. These findings align with previous reports suggesting that mothers may influence transmission of specific taxa to maintain pregnancy and transmit them to their offspring. However, unlike the potentially beneficial genus Bacteroides, which is enriched and transmitted to support offspring survival, further research is needed to understand why Escherichia-Shigella and Fusobacterium, genera with potentially adverse implications, are specifically increased in sows but not in other adult pigs. Higher R² values indicate that differences in microbiota composition account for more variance in microbial community structure within a given niche at a specific time point. The microbiota of piglets at the nursery stage (O2, Day 35) was most similar to that of sows, with a similarity of 92.31%. The lactation stage (O1, Day 5) showed the second highest similarity at 88%. As piglets aged, the similarity decreased, with O3 (Day 80, growing period) at 83.33% and O4 (Day 145, finishing period) at 81.34%. This suggests that maternal seeding plays a key role in early life, but its influence diminishes over time as the microbiota matures with piglet growth. As piglets transitioned from a milk-based diet to a solid feed post-weaning, the composition of gut microbiota changed. Prevotella becomes dominant as piglet transition to solid feed, aiding gut microbiota adaptation by degrading complex carbohydrates. Other genera, such as Blautia, Ruminococcus, Coprococcus, and Treponema, also increase post-weaning, while Bacteroides, Fusobacterium, and Escherichia-Shigella decline. Weaning, the most stressful period for piglets, triggers a major shift in gut microbiota, impacting growth and immunity. Early colonization by Prevotella and Blautia enhances glycan digestion and supports adaptation to a carbohydrate-rich diet. Prevotella also promotes intestinal health through SCFA production and negatively correlates with E. coli infections. As pigs mature, microbial diversity stabilizes, with Prevotella playing a key role in nutrient metabolism, immune modulation, and gut health.
As pigs age, other bacteria, including Ruminococcus, Alloprevotella, Acinetobacter, Myroides, Parabacteroides, Psychrobacter, Rikenellaceae RC9, Treponema, and Sphaerochaeta became more dominant in the gut during the growing-finishing period. During this phase, the significant increase in feed intake and body weight is accompanied by a stable gut microbiota, which helps reduce the risk of intestinal infections while supporting optimal development. However, further research is needed to fully understand the structure and function of the gut microbiota in relation to the host, particularly its physiological, nutritional, and immunological contributions.
The observed differences in α-diversity among lactation, nursery, growing, and finishing pigs, as well as the postpartum stage of the sow, indicate variations in microbial richness, diversity, and evenness across developmental stages. Microbial diversity was lowest on Day 5, likely reflecting the early microbial establishment of the gut microbiota in neonatal piglets. As the pigs matured, microbial richness, diversity, and evenness increased, peaking at Day 80, This trend likely reflects the maturation and stabilization of the gut microbiota over time, as pigs are exposed to a broader range of environmental factors, dietary substrates, and microbial interactions. β-diversity analysis revealed shifts in microbial community composition across the five specific stages, indicating distinct microbial signatures associated with each stage of pig development. These shifts likely result from changes in diet, physiology, and environmental exposure as pig transition between growth stages. Additionally, the identification of stage specific core microbiota provides valuable insights into stable microbial communities and dynamic changes occurring at different stages. While the core microbiome represents a set of microbial taxa consistently present across individuals within a species or population, the stage specific core microbiota consists of microbial communities associated with particular developmental stages. Characterizing core and stage specific microbiota in the swine gut offers potential targets for therapeutic or nutritional interventions aimed at modulating the gut microbiota to improve pig health and productivity at different growth stages.
Our research advances the understanding of the dynamic changes in pig gut microbiota at specific stages and provides insights into bacterial community profiles, aiding in the development of an optimum microbial composition and the identification of potentially beneficial microorganisms.
In this study, we investigated the network of stage specific microbiota to understand the potential functionality of stage specific microbes. We found that the strongest network was built within the sow's microbiome communities, suggesting that the sow is preparing to transfer/seed her microbiome to the offspring. The network of sows immediately after birth is at least three times stronger than that of finishing stage pigs, and up to five times stronger than that of piglets. This data provides potential evidence that the microbiome undergoes significant changes in sows to support pregnancy maintenance and ensuring safe delivery.
Understanding the functional contribution of a microbial community to host physiology and health is crucial for understanding environmental interactions, nutritional processing and metabolism, immune system modulation, pathogen protection, disease prevention and management, development, and growth. We investigated the KEGG based functional metabolic pathways of stage specific microbes using PICRUSt. Functional pathway prediction data comparing S0 and O1 stages revealed that the LACTOSECAT-PWY (lactose and galactose degradation I) was significantly more active in the intestinal microbiota of 5 day old piglets than in sows. Despite both sows and piglets exhibiting a high abundance of Bacteroides, this suggests that the overall microbial community structure and interactions play a crucial role in shaping the functional potential of the gut microbiome, rather than the mere abundance of specific species.
We investigated LEfSe analysis of the KEGG pathways. In newborn piglets (5 day old pigs), metabolic pathways were primarily focused on breast milk and dairy digestion, including LACTOSECAT-PWY (lactose and galactose degradation I), GLUCUROCAT-PWY (superpathway of β-D-glucuronide and D-glucuronate degradation), PWY-6901 (superpathway of glucose and xylose degradation), GALACT-GLUCUROCAT-PWY (superpathway of hexuronide and hexuronate degradation). Energy metabolism and protein synthesis pathways were predicted to be significantly more active in the intestinal microbiota of growing-finishing pigs, as they experience substantial body weight increases during this period.
The LEfSe functional data suggest potential microbiota-host interactions. The LEfSe analysis of functional pathways based on microbial community data confirmed that abundance data alone provide only fragmentary knowledge. To gain a deeper understanding of metabolic functionality, it is necessary to consider complex factors such as the specific functions of each taxon, microbiota networks, and their interactions and relationships. Furthermore, this study investigated the associations between metabolic functional pathways and bacterial genera, as well as the interrelationships among metabolic functional pathways. Through this analysis, we explored how computational microbiome data can provide insights into biologically active pathways. Our findings revealed that the heme biosynthesis pathway, a metabolic pathway closely related to pregnancy maintenance in sows, was predicted to increase in postpartum sows. Moreover, this increase was significantly correlated with the presence of Acinetobacter and Campylobacter, which are known to harbor the HEMESYN2-PWY gene cluster (hemA-hemH). In the O1 stage (lactation stage), Escherichia-Shigella and Enterococcus showed significant associations with the PWY-6629 (superpathway of L-tryptophan biosynthesis), which is directly involved in the metabolic processes responsible for synthesizing L-tryptophan, an essential amino acid required for protein and neurotransmitter production. These findings may provide a potential clue as to why sows transfer Escherichia-Shigella to their piglets. Previous studies have reported that Escherichia-Shigella contributes to tryptophan production, and the biological pathways related to tryptophan are not only crucial for protein synthesis but also play a key role in neurotransmitter biosynthesis. Bacteroides was closely linked to the GLCMANNANAUT-PWY (superpathway of N-acetylglucosamine, N-acetylmannosamine, and N-acetylneuraminate degradation) and Streptococcus exhibited strong correlations with the LACTOSECAT-PWY (lactose and galactose degradation I), which is known to carry the lac operon and is used in dairy fermentation; both pathways have been directly and indirectly implicated in breast milk and dairy digestion. These findings suggest that computational microbiome analysis can provide extensive data on biological activity, enabling prediction and hypothesis generation before conducting wet-lab experiments or on-farm validation studies. Notably, we identified that potentially beneficial genera involved in energy metabolism, such as Butyricicoccus, Parabacteroides, Ruminiclostridium, Ruminococcus, Blautia, Prevotella, Alloprevotella, and Eubacterium were significantly correlated with PWY-7208 (superpathway of pyrimidine nucleobases salvage) and SER-GLYSYN-PWY (superpathway of L-serine and glycine biosynthesis I) which were all enriched in the O3 stage (growing period). Additionally, interpathway correlation analysis revealed that metabolic pathways within each stage were highly interconnected, suggesting that they operate in a coordinated and dynamic manner.
Previous research has investigated the functional characteristics of the swine gut microbiome through metagenomic or predictive methods, identifying important metabolic pathways associated with growth and health. However, these studies were often limited in their ability to longitudinally track maternal microbial transmission and the developmental succession of the microbiome across large and well defined animal cohorts, which are critical for capturing the natural variation and dynamics of microbial communities. Moreover, many of these studies were conducted over a relatively short period, typically covering only the lactation to weaning or nursery stages, and focused solely on piglets rather than on the mother-offspring connection. Our study addresses these limitations by conducting a comprehensive longitudinal analysis involving a substantial cohort of 30 sows and their 179 piglets, allowing us to robustly investigate maternal microbial transmission and the dynamic changes in microbial functions throughout the entire life cycle. This large scale, well characterized cohort provides a more detailed understanding of microbial succession and maternal influences on the swine gut microbiota, offering new insights beyond those available from previous cross sectional studies.
Overall, this study provides foundational computational data on the relationships between metabolic functional pathways and bacterial genera, as well as the interconnections among pathways, offering insights into how well computational microbiome analyses reflect actual biological processes. However, the study relies on PICRUSt2-based functional predictions derived from 16S rRNA data, which, while informative, cannot confirm actual microbial activity. Therefore, future validation using metagenomics and metabolomics is warranted. Moreover, integrating in silico analyses with on farm demonstration trials such as microbiome profiling and blood metabolite analysis following probiotic supplementation in swine production will be crucial for assessing the concordance between computational predictions and real biological activity. Such integration is expected to yield valuable insights for advancing applications in the swine industry.