Genomics analysis of potassium channel genes in songbirds reveals molecular specializations of brain circuits for the maintenance and production of learned vocalizations

Abstract

Background: A fundamental question in molecular neurobiology is how genes that determine basic neuronal properties shape the functional organization of brain circuits underlying complex learned behaviors. Given the growing availability of complete vertebrate genomes, comparative genomics represents a promising approach to address this question. Here we used genomics and molecular approaches to study how ion channel genes influence the properties of the brain circuitry that regulates birdsong, a learned vocal behavior with important similarities to human speech acquisition. We focused on potassium (K-)Channels, which are major determinants of neuronal cell excitability.

Results: We identified 107 K-Channel finch genes, including 6 novel genes common to non-mammalian vertebrate lineages. Twenty human genes are absent in songbirds, birds, or sauropsids, or unique to mammals, suggesting K-Channel properties may be lineage-specific. We also identified specific family members with insertions/deletions and/or high dN/dS ratios compared to chicken, a non-vocal learner. In situ hybridization revealed that while most K-Channel genes are broadly expressed in the brain, a subset is selectively expressed in song nuclei, representing molecular specializations of the vocal circuitry.

Conclusions: Together, these findings shed new light on genes that may regulate biophysical and excitable properties of the song circuitry, identify potential targets for the manipulation of the song system, and reveal genomic specializations that may relate to the emergence of vocal learning and associated brain areas in birds.

Figure 1

Major brain areas for vocal learning and singing in zebra finches. Composite diagram of the songbird brain (parasagittal plane) illustrating the approximate positions and connections of the major nuclei of the song control system. Several related nuclei and connections have been removed for clarity. The song system consists of a direct motor pathway (DMP) for song production (in black) that includes connections from song nucleus HVC to RA, and from RA to brainstem nuclei involved in vocal-motor and respiratory control (nXIIts, RAm, and PAm), and an anterior forebrain pathway (AFP) for song learning (nuclei in grey; projections in white) that includes a cortico-thalmo-cortical loop from Area X in the striatum to thalamic DLM, from DLM to LMAN, and from LMAN back to Area X. Dotted rectangles indicate the approximate positions of the photomicrographs presented in the panels in Figure 4 and Figure 5. See Abbreviations for a complete list of anatomical abbreviations.

Figure 2

Syntenic analysis of novel K-Channel genes (KCNJ3L, KCNE1P, and KCNK16L) that are absent in humans. (A,B and D-F) Schematic representation of conserved chromosomal loci in select vertebrate species that contain orthologs of KCNJ3(A), KCNJ3L(B), KCNJ5/9 L(D), KCNK16(E), and KCNK16L(F). See Results text for details. The relative position of each K-Channel gene of interest is indicated in red; intervening genes are indicated in black. The chromosome or scaffold is indicated beneath each species name. Scale bars indicate the approximate distance between genes. (C) Alignment of predicted amino acid (AA) sequences for zebra finch KCNE1 and KCNE1P. The numbers on the left indicate the relative position of AA residues in each sequence; residues shaded in black are identical, those in gray denote a conservative substitution. The position of highly conserved K-Channel transmembrane domain helices, as well as KCNE Channel signature motifs are indicated in black and grey, respectively.

Figure 3

Syntenic analysis of novel K-Channel genes (KCNV2L, KCNQ1L, and KCTD12L) that are absent in humans. (A-F) Schematic representation of conserved chromosomal loci in select vertebrate species that contain orthologs of KCNV2(A), KCNV2L(B), KCNQ1(C), KCNQ1L(D), KCTD12(E) and KCTD12L(F). See Results text for details. The green box in D highlights the placement of the conserved syntenic group in lizard on Chr_Un. Misplacement was likely caused by a genome assembly error. The relative position of each K-Channel gene of interest is indicated in red; intervening genes are indicated in black. The chromosome or scaffold is indicated beneath each species name. Scale bars indicate the approximate distance between genes.

Figure 4

Differential expression of K-Channel genes in the anterior forebrain pathway (LMAN and Area X) (A). Camera lucida drawing of a parasagittal section (~1.4 mm from the midline) depicting anterior portions of the nidopallium and the medial striatum, including song nucleus LMAN and Area X (approximate location is indicated in Figure 1; laminae are depicted by thin lines; nuclei are indicated by dotted lines). (B-H) Representative photomicrographs of select K-Channel genes that are selectively expressed in LMAN and Area X. While some genes show almost exclusive expression in either LMAN (B) or Area X (E), others show varying patterns of expression in both nuclei, including very specific cell populations. Scale bar: 200 μm. Gene abbreviations are given in Table 1. See Abbreviations for a complete list of anatomical abbreviations.

Figure 5

Differential expression of K-Channel genes in the dorsal thalamus (DLM and Ov). Photomicrographs of in situ hybridizations for KCNA1, KCNAB1, and KCNH8 taken in adjacent parasagittal sections at a brain level that includes song nucleus DLM, and auditory nucleus Ov (approximate location is indicated in Figure 1; dotted lines approximate nuclear boundaries based on dark field illumination; thin lines denote myelinated fibers). K-Channel genes reveal highly selective expression in DLM and Ov (A and B), as well as Ov alone. (C). Scale-bar: 100 μm. See Abbreviations for a complete list of anatomical abbreviations.

Figure 6

Differential expression of K-Channel genes in the direct vocal-motor pathway (HVC and RA). (A, E) Camera lucida drawings of a parasagittal section (~2.0 mm from the midline) depicting the dorso-caudal portion of the nidopallium, including song nucleus HVC (approximate location is indicated in Figure 1; HVC is indicated by arrowheads, ventricle is shaded in grey), and the ventro-caudal portion of the arcopallium and song nucleus RA. The location of a lamina is indicated by the dotted line; arrowheads delineate the approximate boundaries of HVC and RA; thin lines (in E) denote fibers of the occipitomesencephalic (OM) tract. In situ hybridization images for K-Channel genes reveal highly selective (positive or negative) markers of HVC (B-D) and RA (F-H). Scale-bar: 200 μm. See Abbreviations for a complete list of anatomical abbreviations.

Figure 7

Expression profiles for select members of the KCNA and KCNAB gene subfamilies in adult male zebra finch brain. (A) Camera lucida drawing of a parasagittal section (~2.0 mm from the midline). (B-F) Representative photomicrographs of in situ hybridizations of KCNA (B-D) and KCNAB (E-F) subunits that were found to be differentially expressed in one or more nuclei of the song system. LMAN is not consistently present in these sections. See Abbreviations for a complete list of anatomical abbreviations.

Figure 8

Differential expression of KCTD12 and KCTD12L in adult male zebra finch brain. (A-B) Photomicrographs of in situ hybridization for KCTD12(A) and KCTD12L(B). While KCTD12 is differentially expressed in specific nuclei of the song system, as well as other major brain subdivisions, KCTD12L shows relatively low expression throughout the brain with the exception of a specific cell type in the mesopallium. (C) High-power view of representative KCTD12L labeled cells in the mesopallium in a region indicated by the square in panel B. (D-E) High power views of KCTD12 and KCTD23L differential expression in habenula. While KCTD12 shows little to no expression in the habenula (D), KCTD12L labels a very specific population of cells in the medial, but not lateral portions of the habenula (E). Scale bars: 500 μm in A and B; 50 μm in C; 200 μm in D and E. Gene abbreviations are given in Table 1. See Abbreviations for a complete list of anatomical abbreviations.

Figure 9

Differential expression of KCNJ3 and KCNJ3L in adult male zebra finch brain. (A-E) Photomicrographs of in situ hybridizations for KCNJ3 and KCNJ3L. The approximately locations of the photomicrographs in A-C are depicted in the schematic to the right of panel A. The approximate location for the photomicrographs in panels D and E is similar to that for panel C except in a slightly more medial brain section. (A-C)KCNJ3 is highly expressed through the brain in most major cell types, including ependymal cells in the ventricle (A), cells in the granular, Purkinje, and molecular layers of the cerebellum (B), as well as various neuronal and glia populations in brainstem areas and fiber tracts (C). (D-E) In contrast, KCNJ3L shows relatively low to no expression throughout the brain, but very strong expression in neurons of the AVT, and possibly substantia nigra (D; the dotted rectangle indicates the approximate position of the photomicrograph presented in E). (E) High-power view of KCNJ3L labeled cells in AVT. Scale bars: 100 μm in A-C and E; 1 mm in D. Gene abbreviatio ns are given in Table 1. See Abbreviations for a list of anatomical abbreviations.