Songbird Brain Organization and its Molecular Convergence with Humans for Vocal Imitation Learning

Author

Gregory Gedman

Date of Award

2022

Document Type

Thesis

Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Jarvis Laboratory

Abstract

The rich diversity of neural cell populations and their specialized connections give rise to a suite of complex behaviors across vertebrates. The presence of common neural cell types may be a product of either shared ancestry through homology or parallel evolution through convergence. However, the extent as to which these two processes are utilized by diverse nervous systems is poorly understood. Understanding the evolutionary relationships of these neural populations, both within and across vertebrate species, is critical for the study of complex trait evolution. Here, I investigate these two evolutionary processes within the adult avian brain and their impact on our understanding of the evolution of vocal imitation learning across species. Much is still unknown about the number and organization of unique neural cell populations in the avian brain. The avian dorsal and ventral pallium, separated by a vestigial ventricle divide, each contain brain circuits governing complex behaviors like navigation and vocal imitation. However, it is not known whether the cell types in these regions are molecularly distinct or represent homologous populations. In Chapter 1, I tested these competing hypotheses of avian brain organization by investigating potential homology between neural cell populations above and below the ventricle. Using RNA sequencing, I laser capture microdissected (LCM) neural subdivisions and performed differential gene expression and network co-expression analyses. We found that each avian subdivision in the dorsal pallium exhibited remarkable molecular similarity to a region in the ventral pallium. Interestingly, each matching population was defined by the same co-expression networks specialized in anatomical structure development, with such gene regulation similarity being most parsimonious with homologous origins. This work settles a debate of avian brain organization that lasted more than two decades, and it offers important insights into neural cell type diversity and function. With a better understanding of avian neural cell population diversity and functions, I can more accurately assess their homologous or convergent relationships to mammalian neural cell types. In Chapter 2, I investigated the analogous neural cell types involved in vocal imitation between songbirds and humans. Convergent gene expression patterns and connectivity exist between primary motor regions in the human (laryngeal motor cortex, LMC), and songbird (robust nucleus of the arcopallium, RA) controlling the vocal organs. However, it is unknown if the premotor control of these primary motor regions is convergent as well. In songbirds, HVC and LMAN are two premotor regions each critical for proper production and imitation of sounds respectively. In humans, prominent premotor regions activated during speech include the LMC, Broca’s area, and the Supplementary Motor Area (SMA). Using LCM and RNA-Seq, I generated vocal brain region specific gene sets for each of the four principal song nuclei in the zebra finch (Area X, HVC, LMAN, RA) and compared these with genes sets from human brain regions active during speech using gene set enrichment analysis. I found a weak correlation (10 genes) with the genes specialized in zebra finch HVC and human Broca’s area. Similarly, songbird LMAN most closely matched the human SMA (12 genes), with genes enriched for specialized connectivity. In contrast, songbird HVC exhibited the strongest genetic correlation (>200 genes) with human LMC, followed closely by motor RA. Strikingly, utilizing single cell RNA-Seq data from human primary motor cortex (PMC), I found that songbird HVC cells were more like the intratelencephalic neurons in the PMC superficial layers, while songbird RA cells were more like long-range projection neurons in the PMC deep layers. These results offer strong evidence of an analogous microcircuit between two spatially distinct regions in songbirds and two layers of a cortical column in humans for vocal imitation. We molecularly profiled the song system of the budgerigar parakeet to test for further convergence in the parrot lineage, but were limited by sequence gaps and overall quality of the reference genome assembly. To overcome this limitation, I generated a haplotypephased, chromosome-scale budgerigar reference genome assembly that is an order of magnitude more contiguous than the current version, enabling genomic studies for parrot vocal learning and beyond. Collectively, this work has important implications for our understanding of neural cell type evolution both within and across species. My work strongly suggests avian neural populations surrounding the ventricle are homologous, and this more holistic understanding of avian neural cell type homology both allows for a deeper understanding of the evolution of complex behaviors within the avian lineage, and also enables scientists to ask fully informed questions of cross-species cell type comparisons. Looking across lineages, the presence of convergent genetic specializations dictating analogous cell types and circuits in highly divergent species offers new insights into the constraints on the genome to generate the cellular phenotypes necessary for vocal imitation. The existence of analogous microcircuitry utilized by songbirds and humans for vocal production learning confirms the value of the songbird as an invaluable model for speech-motor dysfunction. Overall, this work offers important insights into the evolution of the brain and its remarkable potential to converge on specialized cell types necessary for complex behaviors.

Comments

A Thesis Presented to the Faculty of The Rockefeller University in Partial Fulfillment of the Requirements of the degree of Doctor of Philosophy

Recommended Citation

Gedman, Gregory, "Songbird Brain Organization and its Molecular Convergence with Humans for Vocal Imitation Learning" (2022). Student Theses and Dissertations. 686.
https://digitalcommons.rockefeller.edu/student_theses_and_dissertations/686