Convergent remodelling of the gut microbiome is associated with host energetic condition over long-distance migration

Abstract

The gut microbiome can be thought of as a virtual organ given its immense metabolic capacity and profound effects on host physiology. Migratory birds are capable of adaptively modulating many aspects of their physiology to facilitate long-distance movements, raising the hypothesis that their microbiome may undergo a parallel remodeling process that helps to meet the energetic demands of migration.To test this hypothesis, we investigated changes in gut microbiome composition and function over the fall migration of the Blackpoll Warbler (Setophaga striata), which exhibits one of the longest known autumnal migratory routes of any songbird and rapidly undergoes extensive physiological remodeling during migration.Overall, our results showed that the Blackpoll Warbler microbiome differed significantly across phases of fall migration. This pattern was driven by a dramatic increase in the relative abundance of Proteobacteria, and more specifically a single 16S rRNA gene amplicon sequence variant belonging to the family Enterobacteriaceae. Further, Blackpoll Warblers exhibited a progressive reduction in microbiome diversity and within-group variance over migration, indicating convergence of microbiome composition among individuals during long-distance migration. Metagenomic analysis revealed that the gut microbiome of staging individuals was enriched in bacterial pathways involved in vitamin, amino acid, and fatty acid biosynthesis, as well as carbohydrate metabolism, and that these pathways were in turn positively associated with host body mass and subcutaneous fat deposits.Together, these results provide evidence that the gut microbiome of migratory birds may undergo adaptive remodeling to meet the physiological and energetic demands of long-distance migration.

Keywords: 16S rRNA; Blackpoll Warbler; birds; gut microbiome; metagenomics; migration; passerine; physiology.

Figure 1.. Sampling locations across phases of Blackpoll Warbler fall migration.

(a) Points indicate sampling locations over three phases of Blackpoll Warbler fall migration. Breeding (red points; n = 17): Yukon (Y; n = 2) and British Columbia (BC; n = 15); stopover (orange points; n = 58): Black Swamp Bird Observatory (BSBO; n = 12), Winous Point (WP; n = 8), Powdermill Avian Research Center (PARC; n = 19), and Foreman’s Branch Bird Observatory (FBBO; n = 19); Staging (green points; n = 41): Cape May Bird Observatory (CMBO; n = 15), Block Island (BI; n = 6), Manomet Bird Observatory (MBO; n = 20). (b) Staging adult Blackpoll Warbler (Setophaga striata). Blackpoll Warbler range map provided by BirdLife International and Birds of the World. Blackpoll Warbler photo provided by Manomet Bird Observatory.

Figure 2.. The Blackpoll Warbler gut microbiome exhibited reduced alpha diversity over phases of fall migration.

(a) Chao1 ASV richness (F = 10.25, P < 0.0001) and (b) Shannon diversity (F = 9.67, P = 0.0001) differed significantly across phases of migration: Breeding (n = 17), stopover (n = 58), and staging (n = 41). *** denotes P ≤ 0.001 and * denotes P ≤ 0.05 after Tukey HSD correction.

Figure 3.. Convergence of the Blackpoll Warbler gut microbiome community composition over phases of fall migration.

(a) Phylogenetic RPCA distances differed significantly across phases of migration (PERMANOVA pseudo-F = 16.88, P = 0.001). Ellipses represent 95% confidence level of group centroids. Breeding (red triangles; n = 17): Yukon (Y; n = 2) and British Columbia (BC; n = 15); Stopover (orange squares; n = 58): Black Swamp Bird Observatory (BSBO; n = 12), Winous Point (WP; n = 8), Powdermill Avian Research Center (PARC; n = 19), and Foreman’s Branch Bird Observatory (FBBO; n = 19); Staging (green circles; n = 41): Cape May Bird Observatory (CMBO; n = 15), Block Island (BI; n = 6), Manomet Bird Observatory (MBO; n = 20). (b) Within-group phylogenetic RPCA distances differed significantly across phases of migration (PERMDISP pseudo-F = 5.89, P = 0.004). ** denotes P ≤ 0.01 after FDR correction.

Figure 4.. Remodeling of Blackpoll Warbler gut microbiome composition over phases of fall migration.

(a) Relative abundances of bacterial phyla and (b) families across phases of migration. Letter codes denote sampling locations within phases of migration. Breeding (n = 17): Yukon (Y; n = 2) and British Columbia (BC; n = 15); Stopover (n = 58): Black Swamp Bird Observatory (BSBO; n = 12), Winous Point (WP; n = 8), Powdermill Avian Research Center (PARC; n = 19), and Foreman’s Branch Bird Observatory (FBBO; n = 19); Staging (n = 41): Cape May Bird Observatory (CMBO; n = 15), Block Island (BI; n = 6), Manomet Bird Observatory (MBO; n = 20).

Figure 5.. Taxonomic shifts in the Blackpoll Warbler gut microbiota over phases of fall migration.

The relative abundance of the bacterial phyla (a) Proteobacteria increased while (b) Firmicutes decreased significantly over phases of migration. (c) The relative abundance of the bacterial family Enterobacteriaceae increased significantly over phases of migration. (d) The relative abundance of the bacterial family Enterobacteriaceae increased significantly after the removal of a dominant undescribed ASV. Breeding (n = 17), stopover (n = 58), and staging (n = 41). *** denotes P ≤ 0.001, ** denotes P ≤ 0.01, and * denotes. P ≤ 0.05 after FDR correction.

Figure 6.. The gut microbiome of staging Blackpoll Warblers is enriched in pathways related to energy metabolism.

Log2 fold change plot illustrating 34 differentially enriched bacterial metabolic pathways between breeding (n = 8; red points) and staging (n = 12; green points) Blackpoll Warblers. Red points represent pathways enriched in breeding individuals, while green points represent those enriched in staging individuals. All pathways were significant at the threshold of P ≤ 0.05 after FDR correction.

Figure 7.. The enrichment of bacterial pathways is associated with metrics of Blackpoll Warbler energetic condition.

Log2 fold change plot illustrating 36 unique bacterial metabolic pathways significantly associated with (a) Scaled Mass Index (SMI) and (b) subcutaneous fat scores. Yellow points denote bacterial pathways that were significantly associated with both SMI and fat score. Blue points denote pathways that had a significant positive association with both SMI and fat score. Gray points denote pathways significantly associated with either SMI or fat score. All pathways were significant at the threshold of P ≤ 0.05 after FDR correction.