Now that We’ve Got the Map, Where Are We Going? Moving from Gene Candidate Lists to Function in Studies of Brain Evolution

Abstract

ture microdissection may be cumbersome for acquiring samples at a finer spatial res-olution.T wo recent studies attempt to overcome these limitations by using in situ hybridiza-tion (ISH) to provide new insight into the role of cell type-specific regulation of gene expression in brain evolution [Mashiko et al., 2012; Zeng et al., 2012]. Such compara-tive gene expression atlases usher in a new era of brain mapping, yet they are firmly situated in the venerable old tradition of Brodmann, Vogt, Betz, von Economo, and others by providing detailed descriptions of phenotypic differences in brains across species. Today, however, the toolkit for ob-serving neural phenotypes is much more sophisticated than the few histological stains that were available at the turn of the 20th century. Modern brain mapping stud-ies, which incorporate information from physiology, connectivity, cytoarchitecture, chemoarchitecture, receptor autoradiogra-phy, and gene expression, are essential guides for localizing hotspots of variation, pointing towards the features that make the brains of different species unique. Mashiko et al. [2012] obtained ISH maps of the cerebral cortex, pulvinar, and dLGN from neonatal and fetal common marmosets, a small New World monkey, and compared them to data obtained from The past decade has witnessed a tre-mendous increase in the information available regarding gene expression in the brains of different species. This rapid ac-cumulation of data has been fueled by ad-vances in methodological approaches al-lowing for the efficient detection and mea-surement of mRNA, including techniques such as qRT-PCR, oligonucleotide and cDNA microarrays, RNA-seq, and differ-ent variants of serial analysis of gene ex-pression (SAGE). Although these tech-niques provide a remarkable depth of in-formation about transcriptional activity within a brain region, they are typically performed with homogenates of tissue, which dilutes sensitivity to cell type-spe-cific effects. Thus, major challenges still remain at the intersection of the compara-tive analysis of brain structure and gene expression. Currently, one of the most se-rious obstacles is relating species differ-ences in mRNA transcript levels to the vast complexity of diverse cell types distribut-ed among layers of the neocortex and within subcortical nuclei. One strategy is to use laser capture microdissection to ob-tain samples for genome-wide expression profiling, as has been done for neocortical and dorsal lateral geniculate nucleus (dLGN) layers in rhesus macaques [e.g. Bernard et al., 2012]. However, laser cap-