Vocal practice regulates singing activity-dependent genes underlying age-independent vocal learning in songbirds

Abstract

The development of highly complex vocal skill, like human language and bird songs, is underlain by learning. Vocal learning, even when occurring in adulthood, is thought to largely depend on a sensitive/critical period during postnatal development, and learned vocal patterns emerge gradually as the long-term consequence of vocal practice during this critical period. In this scenario, it is presumed that the effect of vocal practice is thus mainly limited by the intrinsic timing of age-dependent maturation factors that close the critical period and reduce neural plasticity. However, an alternative, as-yet untested hypothesis is that vocal practice itself, independently of age, regulates vocal learning plasticity. Here, we explicitly discriminate between the influences of age and vocal practice using a songbird model system. We prevented zebra finches from singing during the critical period of sensorimotor learning by reversible postural manipulation. This enabled to us to separate lifelong vocal experience from the effects of age. The singing-prevented birds produced juvenile-like immature song and retained sufficient ability to acquire a tutored song even at adulthood when allowed to sing freely. Genome-wide gene expression network analysis revealed that this adult vocal plasticity was accompanied by an intense induction of singing activity-dependent genes, similar to that observed in juvenile birds, rather than of age-dependent genes. The transcriptional changes of activity-dependent genes occurred in the vocal motor robust nucleus of the arcopallium (RA) projection neurons that play a critical role in the production of song phonology. These gene expression changes were accompanied by neuroanatomical changes: dendritic spine pruning in RA projection neurons. These results show that self-motivated practice itself changes the expression dynamics of activity-dependent genes associated with vocal learning plasticity and that this process is not tightly linked to age-dependent maturational factors.

Fig 1. Singing experience–dependent regulation for the sensorimotor learning period of song acquisition in zebra finch.

(A) Critical period of vocal learning (upper panel) and song development (lower panels) in the zebra finch. Blue bars in the lower panels represent the motif structure of crystallized song. Scatter plots indicate the distribution of 500 syllables (duration versus pitch). (B) Examples of songs of SP birds at 1–2 days after release at adult age (100–101 phd). (C) Comparison of syllable acoustic features, pitch, pitch goodness, AM, and entropy variance between normally reared juveniles (n = 5, 40–50 phd), normal adults (n = 5, 100–104 phd), SP birds after release (n = 5, 100–103 phd), and SP birds after song crystallization (n = 5, 119–126 phd). *p < 0.05, **p < 0.01; Welch’s t test. (D) Song development of an SP bird after release at adult age. Blue bars indicate the motif structure of crystallized song. (E, F) Comparison of mean syllable and motif similarities against tutor song between SP birds after release (n = 5, 100–103 phd) and after song crystallization (n = 5, 119–126 phd) and singing-persistent birds (n = 3, 101–105 phd. ***p < 0.005; Welch’s t test. ^NSp > 0.05; unpaired t test. Supporting data can be found in S1 Data. AM, amplitude modulation; NS, not significant; phd, post hatching day; SP, singing-prevented.

Fig 2. Cumulative singing experience regulates a cluster of singing activity–dependent genes in RA.

(A) Schematic showing selected song-control regions and connections in the songbird brain. The posterior motor pathway and the anterior cortical-basal ganglia-thalamic circuit (anterior forebrain pathway) are represented as solid and dotted white lines, respectively. HVC used as a proper name; Area X = Area X of the striatum; nXIIts = tracheosyringeal part of the hypoglossal nucleus. (B) Sampling conditions for RNA-seq to extract transcriptome information on singing experience, age, and singing induction. (C, D) Upper panels: dendrogram of average linkage hierarchical clustering of differentially regulated genes in HVC (C) and RA (D) (3,214 and 1,811 genes, respectively). The red dotted line indicates the height at which the tree was cut. Lower panels: correlation heat maps between gene expression levels and each parameter: singing experience, age, or singing induction. Colored bands indicate positive (green) and negative (red) correlations. (E) Regulation relationships of gene clusters in HVC and RA for singing experience, age, and singing induction. Heat colors show correlations with parameters for each gene cluster. P values in each cell; student’s asymptotic P value. (F) Heat maps of Z scores of RA Cluster I and II genes (119 and 836 genes, respectively), normalized by average expression value of each gene at juvenile silent condition. (G) Cis-enrichment analyses of RA Cluster I and II genes. Each gray bar represents a predicted TF binding site. Supporting data can be found in S2 Data for panels C–G. DLM, dorsal lateral nucleus of the medial thalamus; LMAN, lateral magnocellular nucleus; NIf, interfacial nucleus of the nidopallium; RA, robust nucleus of the arcopallium; RNA-seq, RNA sequencing; SP, singing-prevented; TF, transcription factor.

Fig 3. Direct comparison of the RA transcriptomes under singing condition between SP adult, normal adult, and juvenile groups.

(A) MA plots indicate the expression of differentially regulated RA genes between SP adult singing versus normal adult singing (left) and between SP adult singing versus juvenile singing (right). (B) Number of genes compatibly identified as differentially regulated genes by DEseq2 and RA Cluster I genes by WGCNA. Supporting data can be found in S3 Data. fpkm, fragments per kilobase of exon per million mapped fragments; RA, robust nucleus of the arcopallium; SP, singing-prevented; WGCNA, weighted gene coexpression network analysis.

Fig 4. Cumulative singing experience–regulated gene expression in RA projection neurons between silent and singing condition.

(A) RA Cluster I (Arc, Crem, Nr4a1, Sik1, Dusp5, and Fam60a) and II (Gabra5, Evl, Dpysl3, and Il1rapl2) gene expression in HVC and RA in juvenile, adult, and SP adult (1–2 days after release) birds. Right panels: induction intensity of the singing activity–dependent genes in juvenile (orange), adult (blue), and SP adult (green) birds in HVC and RA. The last 30 minutes of the singing duration of each bird is shown at the bottom. Lines represent linear approximation curve (*p < 0.01, **p < 0.001, ***p < 0.0001, ****p < 0.00001; ANCOVA with Bonferroni correction). (B) Heat maps showing induction differences of Cluster I and II genes in song nuclei between adult, juvenile, and SP adult (1–2 days after release) birds (ANCOVA with Bonferroni correction). Supporting data can be found in S4 Data. ANCOVA, analysis of covariance; DLM, dorsal lateral nucleus of the medial thalamus; LMAN, lateral magnocellular nucleus; NIf, interfacial nucleus of the nidopallium; RA, robust nucleus of the arcopallium; SP, singing-prevented.

Fig 5. Singing experience–mediated dendritic spine pruning in RA projection neurons.

(A) Coinduction of RA cluster I genes (Arc, Nr4a1, Sik1, and Dusp5) after juvenile singing in glutamatergic neurons with Vglut2 (+) but not GABAergic neurons with Gad2 (+). Filled and empty arrowheads: cells that coexpressed or did not coexpress with singing activity–dependent genes and cell marker genes, respectively. Cell nuclei (blue, DAPI). Scale bar = 40 μm. Bar graphs: proportion of each subpopulation in cells that express the mRNA of RA Cluster I genes. (B) Selective induction of Arc mRNA (green) after juvenile singing in RA projection neurons. Diagram of DiI retrograde labeling (red) of RA projection neurons to nXIIts. Cell nuclei (blue, DAPI). (C) Golgi-stained RA projection neurons and RA-surrounding arcopallial neurons in juvenile (55 phd), adult (105 phd), and SP adult (101 phd at 1–2 days after release) birds. Scale bars = 5 μm (upper) and 50 μm (lower). Bar graphs: dendritic spine density of RA projection neurons and RA-surrounding arcopallial neurons in juveniles (n = 18 cells from 6 birds), adults (n = 15 cells from 5 birds), and SP adults. ***P < 0.0001, NS: p > 0, unpaired t test with Bonferroni correction. Error bars: SEM. Supporting data can be found in S5 Data. NS, not significant; phd, post hatching day; RA, robust nucleus of the arcopallium; SP, singing-prevented.