Impact of early-life feeding on local intestinal microbiota and digestive system development in piglets

Abstract

Early-life gut microbial colonisation is known to influence host physiology and development, shaping its phenotype. The developing gastro-intestinal tract of neonatal piglets provides a "window of opportunity" for programming their intestinal microbiota composition and corresponding intestinal development. Here, we investigated the impact of early feeding on jejunum and colon microbiota composition, and intestinal maturation in suckling piglets. From two days of age, early-fed (EF; n = 6 litters) piglets had access to solid feed containing a mixture of fibres till weaning (day29) in addition to sow's milk, whereas the control (CON; n = 6 litters) piglets exclusively fed on sow's milk. Early feeding elicited a significant impact on the colon microbiota, whereas no such effect was seen in the jejunal and ileal microbiota. Quantified eating behavioural scores could significantly explain the variation in microbiota composition of EF piglets and support their classification into good, moderate, and bad eaters. Members of the Lachnospiraceae family, and the genera Eubacterium, Prevotella, and Ruminococcus were quantitatively associated with eating scores. EF piglets were found to have a decreased pH in caecum and colon, which coincided with increased short-chain fatty acid (SCFA) concentrations. Moreover, they also had increased weights and lengths of several intestinal tract segments, as well as a decreased villus-crypt ratio in jejunal mucosa and an increased abundance of proliferative cells in colon mucosa. The approaches in this study indicate that early feeding of a mixed-fibre (pre-weaning) diet changes the microbiota composition, pH, and fermentation products in the distal gut of piglets, while it also alters both macroscopic and microscopic intestinal measurements. These results exemplify the potential of early feeding to modulate intestinal development in young piglets.

Figure 1

Microbiota composition along the intestinal tract. (A) Principal component analysis of jejunum, ileum and colon microbiota (PC1 = 66.66%, PC2 = 4.76%) at genus level. (B) Redundancy analysis (explained variation = 67.8% P = 0.002) of intestinal location with associated microbial groups at family level. Microbial groups visualized in this figure have a minimum fit value of at least 40% on the horizontal axis and a response score > 0.63 and > 0.80 for jejunum/ileum and colon, respectively. Specific microbial families which are differentially abundant in jejunum/ileum were enforced in this graph). (C) Alpha diversity (Chao1 bias corrected) comparison among jejunum, ileum and colon. Significant differences between groups were assessed by student t test or Mann–Whitney U test (***: P < 0.001).

Figure 2

Colonic microbiota composition in early-fed (EF) and control (CON) group. (A) Redundancy analysis at genus level (PC1 = 8.75%, PC2 = 16.12%; P = 0.002) with associated microbial groups shown (minimum fit value of at least 30% and > 0.55 response score on horizontal axis). (B) Alpha diversity (Chao1 bias corrected) comparison between the two groups. (C) Heat map showing relative abundance of discriminative bacterial genera (≥ 0.40 response score in x axis) as found in redundancy analysis. RDA identified microbes that were also detected in a previous study with similar design are shown in bold.

Figure 3

Classification of early-fed (EF) piglets into good (green), moderate (blue) and bad (red) eaters. (A) Individual piglet’s eating behaviour quantification (14 EF piglets; total eating seconds per week) for four weeks pre-weaning by video observation. (B) Good (> 2 × median; green), moderate (between 0.8 × and 2 × median; blue) and bad (below 0.8 × median; red) eaters, grouping based on eating observed in the “last two days” before weaning. (C) Redundancy analysis based on eating scores from “last two days” before weaning (explained variation = 5.42%, P = 0.024), establishing the microbiota discrimination between the “good”, “moderate” and “bad” classification within the EF piglets (minimum fit value of at least 30% and > 0.55 response score on horizontal axis). (D) Spearman correlation of individual microbial genera with the eating score from “last two days”. Lachnospiraceae ND3007, Eubacterium xylanophilum, Prevotella 1 and Odoribacter were identified in redundancy analysis.

Figure 4

Impact of early feeding on pH and short chain fatty acids (SCFAs). (A) Digesta pH in different parts of the GIT for EF and CON and an expansion of the two segments where significant difference in pH was detected, illustrating the distribution of individual piglets in both groups (using the EF group classification of good (green), moderate (blue) and bad (red) eaters; see Fig. 2B). (B) Correlation of pH caecum/colon of individual EF piglets with the quantified eating score from last two days (Caecum: r =  − 0.48, P = 0.087; Colon: r =  − 0.33, P = 0.24). (C) Group level comparison for colonic SCFA concentration (µmol/g wet weight of digesta) in piglets. Significant differences between groups were assessed by student t test or Mann–Whitney U test (*: P < 0.05; **: P < 0.01; ***: P < 0.001). EF = early-fed group; CON = control group. Green = good eaters; Blue = moderate eaters; Red = bad eaters.

Figure 5

Effects of early feeding on (A) intestinal morphometry and (B) PCNA⁺ proliferating cells in jejunum and colon (representative image) at weaning (day29). Hoechst and PCNA positive cells are represented in blue- and green-coloured cells respectively. EF = early-fed group; CON = control group. Differences between groups were assessed by student t test or Mann–Whitney U test (*, P < 0.05). The bars in these images illustrate the measured parameters. Representative images of CON and EF groups can be found in Supplementary Fig. 11.