Effects of Feather-Pecking Phenotype on Physiological and Neurobiological Characteristics and Gut Microbiota Profile of Goslings

Abstract

FP is a detrimental behavior for chickens, ducks, and geese associated with numerous physiological and neurobiological characteristics, which have been identified in many species as regulated by the gut microbiota. However, it is unknown whether and how gut microbiota influences FP by regulating neurotransmitter systems in geese. This study aimed to investigate the phenotypic correlation between feather pecking and changes in physiological, neurobiological, and gut microbiota profiles in gosling. Three behavioral phenotypes were observed in goslings, including severe feather peckers (SFPs), victims of SFPs, and non-peckers (NFPs). The significantly lower feather scores and body weights were observed in victims compared to both SFPs and NFPs (p < 0.05). Regarding the physiological phenotype, victims had higher dopamine (DA) levels than NFPs, and SFPs had lower 5-hydroxytryptamine (5-HT) in the serum than NFPs (p < 0.001), with intermediate 5-HT levels in victims. Victims had lower glutathione peroxidase (GSH-Px) compared to SFPs and NFPs (p < 0.05). Moreover, higher mRNA expression levels of HTR1A, SLC6A4, and TPH2 in the 5-HT metabolic pathway were detected in NFPs than those in SFPs and victims (p < 0.05). In addition, regarding gut microbiota measured by 16S rRNA sequencing, SFPs had lower diversity and comparable cecal microbiota compared to victims and NFPs. Proteobacteria, Verrucomicrobia, Ruminococcus spp., and Bilophila spp. were enriched in SFPs, while Bacteroides and Parabacteroides were enriched in NFPs. From the predicted bacterial functional genes, the cAMP signaling pathway, cGMP-PKG signaling pathway, and pyruvate metabolism were activated in SFPs. The correlation analysis revealed that the genera Bacteroides spp. were associated with differences in 5-HT metabolism between the SFPs and NFPs. In summary, differences in the cecal microbiota profile and 5-HT metabolism drive FP phenotypes, which could be associated with the reduced gut abundance of the genera Bacteroides spp.

Keywords: 5-HT; feather pecking; gosling; microbiota–gut–brain axis.

Figure 1

The physical characteristics of the gosling classification. N = 30. (A) Three categories of FP goslings: severe feather peckers (SFPs), victims of SFP (victims), and non-peckers (NFPs). (B) Feather scores. (C) Body weight. (D) Feather residue from gizzard contents. (E) Length of the cecum. (F) Adrenal index. * p < 0.05, ** p < 0.01.

Figure 2

Detection of indicators of serum antioxidant capacity and physiological stress. (A) Comparison of serum antioxidant indices of MDA, SOD, and GSH-Px. (B) Comparison of hormones E and CORT in goslings. * p < 0.05, ** p < 0.01.

Figure 3

Detection of 5-HT and DA in serum. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 4

Differential expression of 5-HT metabolism-related genes ((A) TPH2, (B) HTR1A, and (C) SLC6A4). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

16S rRNA analysis of the gut microbiome of three groups of goslings. (A) Sparse curve of intestinal flora. (B) Venn diagrams were used to represent the operational taxonomic units (OTUs) observed in three groups of goslings. (C,D) Estimates of abundance indices at the phylum and genus level for three groups. (E) Estimates of abundance indices at the phylum level for three groups of goslings.

Figure 6

Diversity analysis of gut microbiota in gosling. (A) Gut microbiota alpha diversity index in goslings of SFP, victim, and NFP groups. (B) The principal coordinates (PCA). * p < 0.05, ** p < 0.01.

Figure 7

An analysis of the linear discriminant analysis effect size was performed on the gut microbiota of goslings. (A) Circos sample species relationship map demonstrating the distribution of microbial species present in the microflora of three groups of goslings. (B) Bacterial composition comparisons among the groups on phylum level (Kruskal–Wallis H test). (C) LDA scores (LDA > 3) of taxonomic units showing differences in the abundance of SFPs, victims, and NFPs. Red, blue, and green bars correspond to taxonomic units enriched in SFPs, victims, and NFPs, respectively. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 8

Integration of random forest analysis and PICRUSt2-based functional prediction. (A,B) PICRUSt2 predicts functional information about microbial communities in environmental samples. (C) Random forest analysis. (D) Bar graph of COG functional classification statistics. * p < 0.05, ** p < 0.01.

Figure 9

Correlation between gut flora genus, feather score, physiological indices, and antioxidant capacity. (A) Spearman correlation coefficients of the correlation between gut flora genus, feather score, physiological indices, and antioxidant capacity with different bacterial genera. (B) Correlation between feather score, physiological indices, and antioxidant capacity with Bacteroides, Bilophila, and Odoribacter. The heatmap shows statistically significant correlation values (r). Red squares indicate significant positive correlation (0 < r ≤ 1), white squares indicate no correlation (r = 0), and blue squares indicate significant negative correlation (1 ≤ r < 0); darker colors indicate greater correlation. * p < 0.05, ** p < 0.01, *** p < 0.001. (C) Schematic diagram of the gut microbiota–antioxidant indicators–hormone axis.