Urotensin-related gene transcripts mark developmental emergence of the male forebrain vocal control system in songbirds

Abstract

Songbirds communicate through learned vocalizations, using a forebrain circuit with convergent similarity to vocal-control circuitry in humans. This circuit is incomplete in female zebra finches, hence only males sing. We show that the UTS2B gene, encoding Urotensin-Related Peptide (URP), is uniquely expressed in a key pre-motor vocal nucleus (HVC), and specifically marks the neurons that form a male-specific projection that encodes timing features of learned song. UTS2B-expressing cells appear early in males, prior to projection formation, but are not observed in the female nucleus. We find no expression evidence for canonical receptors within the vocal circuit, suggesting either signalling to other brain regions via diffusion or transduction through other receptor systems. Urotensins have not previously been implicated in vocal control, but we find an annotation in Allen Human Brain Atlas of increased UTS2B expression within portions of human inferior frontal cortex implicated in human speech and singing. Thus UTS2B (URP) is a novel neural marker that may have conserved functions for vocal communication.

Figure 1

Syntenic relationships of urotensin receptor genes in birds, based on updated annotations from Table 1. (a) Chicken and zebra finch UTS2R1 and UTS2R5 are syntenic on chromosome 18. (b) UTS2R2 is present on chromosome 14 in chicken. The syntenic locus is annotated as a pseudogene in falcons, and completely absent in zebra finch, great tit, Bengalese finch, and budgerigar. (c) UTS2R3 is present and syntenic in chicken, great tit, and canary, but the syntenic loci in zebra finch, Bengalese finch, and collared flycatcher are disrupted by truncations or deletions and do not predict functional 7-transmembrane domain proteins (Fig. S1).

Figure 2

Brainstem UTS2B and UTS2 gene expression. All photomicrographs are representative of male or female sagittal sections (no sex differences detected, number of observations is given in Table S1), showing UTS2B labelling (except UTS2 in panel i). (a) The nucleus of Edinger-Westphal (EW), the dorsal part of the nucleus of the oculomotor nerve (n3d), the ventral part of the nucleus of the oculomotor nerve (n3v), the nucleus of the trochlear nerve (n4), the red nucleus (Ru); (b) The principal sensory nucleus of the trigeminal nerve (n5p), several unnamed pons nuclei; (c) the magnocellular nucleus, MC; (d) The tracheosyringeal motor nucleus (nXIIts), the supraspinal nucleus (SSp); (e) The nucleus of the abducens nerve, n6; (f) Scattered in the pons; (g) the parvocellular part of the isthmic nucleus, Ipc; (h) the nucleus of the facial nerve, n7; (i) UTS2 expression in the pons; (j,k) Comparison of large UTS2B labelled cells in nXIIts [j] and smaller cells in HVC [k]. Scale bar = 250 μm (a–i) or 25 µm (j,k).

Figure 3

Male Pallium UTS2B gene expression. (a) In situ hybridization image of lateral sagittal section showing labelled cells scattered throughout the pallium, with concentrations of labelled cells dorsally in HVC and in a caudal portion of the arcopallium excluding RA (rostral to the right, dorsal up). (b) Line drawing of section in panel a labelling major landmarks (A – arcopallium; H – hyperpallium; LSt – lateral striatum; M –mesopallium; N – nidopallium; the lateral ventricle indicated by thick line running adjacent to caudal and dorsal borders); dashed lines indicate insets shown at higher magnification in panels c and d. (c) High UTS2B gene expression within cells localized in HVC. (d) high UTS2B gene expression within cells localized to a caudal portion of the dorsal arcopallium (A), adjacent to the robust nucleus of the arcopallium (RA). Arrowheads point to the pallial/sub-pallial lamina (LPS) running just dorsally to the area of cellular labelling. Number of observations is given in Table S1; scale bar = 500 μm.

Figure 4

URP detection by immunohistochemistry. (a) Parasagittal section of caudal telencephalon showing increased immunostaining for URP in an oval area corresponding to nucleus HVC, immediately beneath the lateral ventricle. (b) Box inside HVC enlarged from panel a, showing cytoplasmic immunostaining pattern of a subset of cells (small inset, Hoechst counterstaining revealing all cell nuclei). (c) Field inside Area X, with negligible immunostaining. (d) Absence of primary antibody demonstrates negligible non-specific fluorescence (field inside HVC as in b). (e) Large labelled cells in brainstem nucleus nXIIts. (f) Large labelled cells in brainstem supraspinal nucleus (SSp). (g) Intense puncta of URP staining suggestive of synaptic terminals in RA. Bar = 100 µm (a) or 50 µm (b–g).

Figure 5

Selective expression of UTS2B in RA-, but not X-projecting HVC neurons. Schematics on the left depict injections of fluorescent tract-tracer (CTB-Alexa 488) into RA (top) or Area X (bottom), resulting in retrograde labelling of specific populations of projection neurons in HVC. (a) Retrogradely-labelled RA-projecting neurons (a₁, green) co-localize with UTS2B-expressing cells (a₂, blue), as confirmed by the merged image (a₃). (b) Retrogradely-labelled X-projecting neurons (b₁, green) do not co-localize with UTS2B-expressing cells (b₂, blue), as confirmed by the merged image (b₃). In all panels, arrowheads indicate location of individual backfilled neurons identified by characteristic donut-shaped cytoplasmic labelling that surrounds a nucleus (red, propidium iodide stain). UTS2B expression is revealed by fluorescent in situ hybridization with antisense riboprobes.

Figure 6

UTS2R5 gene expression. In situ hybridization analysis. (a) Key for the figure, showing a schematic projection of labelled brain regions onto a 2-dimensional medial sagittal plane (note that these regions are distributed across multiple planes in the third medial-lateral dimension, not represented here); (b) labelling across the hippocampus (HP); (c) dorsal nucleus of the hyperpallium (DNH); (d) ventral medial arcopallium, rostral (AMVr); (e) substantia nigra (SN); (f) principal sensory nucleus of the trigeminal nerve (n5p); (g) the parvocellular part of the isthmic nucleus (Ipc); (h) the lateral nucleus of the cerebellum (CbL); (i–l) All four photomicrographs are from the same male that showed no visible UTS2R5 gene expression in [i] HVC, [j] RA, [k] area X, while having [l] high UTS2B gene expression in HVC. Photomicrographs [b–h] are representative of males and females (no sex differences detected; number of observations is given in Table S1). Scale bar = 250 μm.

Figure 7

Developmental HVC UTS2B gene expression. Representative low power photomicrographs of the caudal forebrain in females (columns 1 and 2) and males (columns 3 and 4). Columns 2 and 4 show high power insets as indicated. Number of observations is given in Table S1; arcopallium – A; mesopallium - M; nidopallium - N; caudolateral nidopallium – NCL. Scale bars = 500 μm.