The genome of a songbird

Abstract

The zebra finch is an important model organism in several fields with unique relevance to human neuroscience. Like other songbirds, the zebra finch communicates through learned vocalizations, an ability otherwise documented only in humans and a few other animals and lacking in the chicken-the only bird with a sequenced genome until now. Here we present a structural, functional and comparative analysis of the genome sequence of the zebra finch (Taeniopygia guttata), which is a songbird belonging to the large avian order Passeriformes. We find that the overall structures of the genomes are similar in zebra finch and chicken, but they differ in many intrachromosomal rearrangements, lineage-specific gene family expansions, the number of long-terminal-repeat-based retrotransposons, and mechanisms of sex chromosome dosage compensation. We show that song behaviour engages gene regulatory networks in the zebra finch brain, altering the expression of long non-coding RNAs, microRNAs, transcription factors and their targets. We also show evidence for rapid molecular evolution in the songbird lineage of genes that are regulated during song experience. These results indicate an active involvement of the genome in neural processes underlying vocal communication and identify potential genetic substrates for the evolution and regulation of this behaviour.

Figure 1. Divergent patterns of dosage compensation in birds

a, b, The male to female (M/F) ratio of gene expression, measured by species-specific microarrays, is plotted along the Z chromosome of chicken (a) and zebra finch (b). Each point represents the average M/F ratio of a sliding window of 30 genes plotted at the median gene position and stepping one gene at a time along the chromosome. Note region of lower M/F ratios in chicken surrounding the locus of the MHM (male hypermethylated) ncRNA. In zebra finch, genes adjacent to the comparable MHM position (asterisk) show no special cluster of dosage compensation (low M/F ratios), and no MHM sequence appears in the genome assembly. bp, base pairs.

Figure 2. Enriched expression of a CR1-like element in the zebra finch song system

a, Genomic alignment of an RNA containing a CR1-like retrotransposon element (in blue) and adjacent ESTs, with respective GenBank accession numbers. b–d, DV949717 is expressed in the brain of adult males with enrichment in song nuclei HVC (letter-based name) and LMAN (lateral magnocellular nucleus of the anterior nidopallium), as revealed by in situ hybridization. The diagram in b indicates areas shown in photomicrographs in c and d. Cb, cerebellum; Hp, hippocampus; Meso, mesopallium; Nido, nidopallium; Shelf, nidopallial shelf region; St, striatum. Scale bars, 0.1 mm.

Figure 3. miR-124 in the auditory forebrain is suppressed by exposure to new song

TaqMan assays comparing samples from the auditory lobule of adult male zebra finches in silence (open bars) or 30 min after onset of new song playback (filled bars). a, Comparison of two sample pools, each containing auditory forebrains of 20 birds. b, Comparisons of paired individual subjects, n = 6 pairs (P = 0.03, Wilcoxon paired test). Error bars denote s.e.m. of triplicate TaqMan assays. Parallel TaqMan analyses of the small RNA RNU6B were performed with all samples and showed no significant effect of treatment for this control RNA.

Figure 4. Conserved NR4A3 3′ UTR is a potential region for microRNA integration

a, zPicture alignment of 3′ portion of zebra finch to human gene showing UTR region of high similarity beyond the coding exons. Dark red bars, regions with the highest sequence conservation; black rectangles, position of song-regulated ESTs within the conserved UTR but outside the Ensembl gene model (ENSTGUG00000008853). b, Alignment of zebra finch and human 3′ UTR sequences showing the per cent sequence identity for each evolutionarily conserved region. Dots indicate positions of conserved new (‘n-’) or established (‘miR-’) microRNA-binding sites in both species within these regions.

Figure 5. Transcriptional control network in area X engaged by singing

a, Clustered (1–20) temporal expression profiles of 807 genes (rows) that change with time and amount of singing; red, increases; blue, decreases; white, no change relative to average 0-h control. Grey/coloured bars on left, clusters with enrichment of specific promoter motifs (P < 0.01). b, Enriched transcription-factor-binding motifs (abbreviations) found in the promoters of late response genes, clusters 9–12 (coloured as in a); bold, binding sites for known activity-dependent transcription factors (for example, CREBP1) or transcription factor complexes (for example, CREBP1–CJUN); black, sites for post-translationally activated transcription factors; brown, sites for transcriptionally activated transcription factors including by singing (for example, in cluster 1). Graph shows time course of average expression of all genes in the late response clusters, normalized to average 0 h for that cluster. Also plotted is the average expression of the C-FOS transcription factor mRNA, which binds to the AP-1 site over-represented in the promoters of cluster 10 genes.

Accessory Figure 1. Predicted motif genome location distribution

(A) Density of predicted motifs (y axis) for different categories of regions (x axis) in terms of location with respect to their nearest genes, shown in blue. The horizontal pink line is the genome-wide average. See text for definitions of motif density and various region categories. (B) P-values of enrichment or depletion of motif occurrence in each category of regions, using one-tailed Fishers exact tests. Negative logarithms are shown.

Accessory Figure 2. Motif target counts by defined location

Counts of targets windows of individual motifs in different categories of regions are compared to the respective expected values, with colors shown indicating whether the count is greater or less than expectation. Green cells correspond to counts that are higher than the average and red cells correspond to regions with below-average counts.

Accessory Figure 3

Comparative analysis of marker order on chicken chromosomes 2-8 and Z (GGA2-8, GGAZ) and their zebra finch orthologues (TGU2-8, TGUZ). The central part of each figure was created by aligning whole chromosomal sequences using the program GenAlyzer. Line colour indicates the length of sequences with 100% sequence identity. The tentative chromosomal rearrangements suggested by this analysis were verified using fluorescent in situ hybridization (FISH). Letters indicate the position of chicken and zebra finch BACs with orthologous sequence content in the genome sequences of both species (see accessory file Physical mapping table 2009-09-16.xls for details on the FISH probes used). Red dots on the ideograms illustrate the physical chromosomal position as determined by FISH.

Accessory Figure 3

Comparative analysis of marker order on chicken chromosomes 2-8 and Z (GGA2-8, GGAZ) and their zebra finch orthologues (TGU2-8, TGUZ). The central part of each figure was created by aligning whole chromosomal sequences using the program GenAlyzer. Line colour indicates the length of sequences with 100% sequence identity. The tentative chromosomal rearrangements suggested by this analysis were verified using fluorescent in situ hybridization (FISH). Letters indicate the position of chicken and zebra finch BACs with orthologous sequence content in the genome sequences of both species (see accessory file Physical mapping table 2009-09-16.xls for details on the FISH probes used). Red dots on the ideograms illustrate the physical chromosomal position as determined by FISH.

Accessory Figure 3

Comparative analysis of marker order on chicken chromosomes 2-8 and Z (GGA2-8, GGAZ) and their zebra finch orthologues (TGU2-8, TGUZ). The central part of each figure was created by aligning whole chromosomal sequences using the program GenAlyzer. Line colour indicates the length of sequences with 100% sequence identity. The tentative chromosomal rearrangements suggested by this analysis were verified using fluorescent in situ hybridization (FISH). Letters indicate the position of chicken and zebra finch BACs with orthologous sequence content in the genome sequences of both species (see accessory file Physical mapping table 2009-09-16.xls for details on the FISH probes used). Red dots on the ideograms illustrate the physical chromosomal position as determined by FISH.

Accessory Figure 3

Comparative analysis of marker order on chicken chromosomes 2-8 and Z (GGA2-8, GGAZ) and their zebra finch orthologues (TGU2-8, TGUZ). The central part of each figure was created by aligning whole chromosomal sequences using the program GenAlyzer. Line colour indicates the length of sequences with 100% sequence identity. The tentative chromosomal rearrangements suggested by this analysis were verified using fluorescent in situ hybridization (FISH). Letters indicate the position of chicken and zebra finch BACs with orthologous sequence content in the genome sequences of both species (see accessory file Physical mapping table 2009-09-16.xls for details on the FISH probes used). Red dots on the ideograms illustrate the physical chromosomal position as determined by FISH.

Accessory Figure 3

Comparative analysis of marker order on chicken chromosomes 2-8 and Z (GGA2-8, GGAZ) and their zebra finch orthologues (TGU2-8, TGUZ). The central part of each figure was created by aligning whole chromosomal sequences using the program GenAlyzer. Line colour indicates the length of sequences with 100% sequence identity. The tentative chromosomal rearrangements suggested by this analysis were verified using fluorescent in situ hybridization (FISH). Letters indicate the position of chicken and zebra finch BACs with orthologous sequence content in the genome sequences of both species (see accessory file Physical mapping table 2009-09-16.xls for details on the FISH probes used). Red dots on the ideograms illustrate the physical chromosomal position as determined by FISH.

Accessory Figure 4. Bayesian phylogenies of zebra finch MHC genes

(A) Class I genes and (B) Class IIB genes were compared to sequences from the chicken MHC-B complex. Putatively functional zebra finch genes with open reading frames are given numerical suffixes and putative pseudogenes are given lettered suffixes. For Class I we also include a chicken sequence from the MHC-Y region (YFV). For zebra finch class I, we show the placement of eight brain ESTs (indicated by their GenBank accession numbers) supporting the expression of MHC Class I genes in the brain. Posterior probabilities are given for well-supported nodes in the tree.

Accessory Figure 5

General view showing WGAC (>5kb) and WSSD on all chromosomes. Grey above lines is WSSD and red below lines is WGAC. ChrUn was treated as a “distinct” chromosome.

Accessory Figure 6

(A) Characteristics of co-expressed gene sets from Dong et al. ⁵⁸ (See S3 supplementary notes). “Gene set”: name of the gene set, as in the original paper. “All genes” refers to the genes on the array in Dong et al. Numbers in parentheses indicate component subsets of a set. “Size”: number of genes in set. “terr_len”: average gene territory length of a gene set. “gene_len”: average coding sequence length. “intergenic_length”: average of (territory length – gene length). “p-value”: statistical significance of enrichment for short (pink cells) or long (green cells) territories, as measured by 2-tailed Wilcoxon Rank Sum tests. (B) Average gene length and intergenic length of a gene set (y axis) versus average gene territory length. Each point corresponds to a gene set.