Adaptive Regulation of Stopover Refueling during Bird Migration: Insights from Whole Blood Transcriptomics

Abstract

Migration is one of the most energy-demanding tasks in avian life cycle. Many birds might not have sufficient fuel stores to cover long distances, so they must stop to rest and refuel at stopover sites, especially after the crossing of large ecological barriers. There, birds undergo several behavioral, morphological, and physiological trait adjustments to recover from and prepare for their journey; however, regulation of such processes at the molecular level remains largely unknown. In this study, we used transcriptomic information from the whole blood of migrating garden warblers (Sylvia borin) to identify key regulatory pathways related to adaptations for migration. Birds were temporarily caged during spring migration stopover and then sampled twice at different refueling states (lean vs. fat), reflecting different migratory stages (stopover arrival vs. departure) after the crossing of an extended ecological barrier. Our results show that top expressed genes during migration are involved in important pathways regarding adaptations to migration at high altitudes such as increase of aerobic capacity and angiogenesis. Gene expression profiles largely reflected the two experimental conditions with several enzymes involved in different aspects of metabolic activity being differentially expressed between states providing several candidate genes for future functional studies. Additionally, we identified several hub genes, upregulated in lean birds that could be involved in the extraordinary phenotypic flexibility in organ mass displayed by avian migrants. Finally, our approach provides novel evidence that regulation of water homeostasis may represent a significant adaptive mechanism, allowing birds to conserve water during long-distance flight, mainly through protein catabolism.

Keywords: RNA-seq; candidate genes; differential gene expression; metabolism; phenotypic plasticity; water homeostasis.

Fig. 1.

Phenotype and overall gene expression patterns in the whole blood of garden warblers. (A) Boxplot showing body mass measurements of individuals belonging in the two condition groups explored (lean vs. fat). Lines connect each paired sample (samples from the same individual). (B) Heat map showing relationships among samples based on all expression data. (C) Principal components (PC1 and PC2) for the gene expression data after VST. (D) Graphical representation of the clustering of 50 most expressed genes.

Fig. 2.

Transcriptome-wide differential gene expression analyses. (A) Heat map of 1,167 DEGs between the two refueling states. Scaled expression values show upregulated (positive z-score) and downregulated (negative z-score) genes. Dendrogram groups correspond to fat (left) and lean (right) refueling states. (B) Volcano plot of the distribution of all DEGs, depicting the log2fold change and negative log10 nominal P-value for all the expressed genes. Points colored green, red, and blue indicate genes with an adjusted P > 0.05, P > 0.05, and P < 0.05, respectively. (C) The top enriched KEGG pathways for DEGs. Lollipop diagrams show fold enrichment, significance (False Discovery Rate in log10), and number of genes in each pathway. The P-values are adjusted using Benjamini–Hochberg correction and the cutoff is 0.05. (D) Boxplots of DEGs (upregulated in lean state, top half; upregulated in fat state, bottom half) in the whole blood of garden warblers with the highest fold changes in expression between states.

Fig. 3.

WGCNA to identify hub genes and pathways associated with different refueling states. (A) Dendrogram depicting clustering of genes based on 1-Topological Overlap (1-TO) distance. (B) Module-refueling state correlation barplot. Significant modules are indicated with asterisks. (C) Boxplots of relative ME expression between lean and fat states in black, brown, and green modules from left to right. (D) Coexpression network of hub genes (showing only genes with weighted connection >0.4 for clarity) upregulated in the lean state (black and brown modules) and upregulated in the fat state (green module).