Things Change: How Comparative Transcriptomics Suggest the Pallium Has Evolved at Multiple Levels of Organization

Despite a century of investigation [Smith, 1919; Holmgren, 1922; Johnston, 1923; Northcutt and Kaas, 1995; Karten, 1997; Butler and Hodos, 2005; Striedter, 2005], the evolutionary origins of the cerebral cortex remain intensely disputed [Karten, 1997; Wang et al., 2010; Suzuki et al., 2012; Dugas-Ford et al., 2012]. As brain tissue does not fossilize, comparative neurobiologists infer ancestral forms from living animals. Unfortunately, one of the earliest branches of the mammalian lineage, Eutriconodonta, went extinct at the end of the Cretaceous period (~145 to ~66 Mya). Thus, to gain insights into the evolutionary origin of the isocortex, which already existed in our most recent common ancestor with monotremes, we need to compare living sauropsids (birds and reptiles) and mammals, two clades separated by more than 300 million years of independent evolution in diverse ecological niches. Numerous studies have noted similarities in connectivity and gene expression between the layers of the mammalian isocortex and nuclear structures of the avian pallium [Karten, 1997; Dugas-Ford et al., 2012]. However, these similarities are seemingly inconsistent with homologies inferred from morphogenetic developmental data. Furthermore, gene expression has been used to support a wide variety of mutually contradictory homologies between isocortical laminae and various sectors of the avian pallium. It is in this context that we aimed to conduct an unbiased gene expression study using an objective analytical method [Belgard et al., 2013]. Because adult brain regions differ in their transcriptomes, we dissected several structures of both controversial and non-controversial homology in adult chickens and mice, and then sequenced and compared their transcriptomes. Although we would have liked to look throughout development, young embryonic brain regions contain so little RNA that this approach was not feasible with the then available technology. Therefore, we instead sequenced RNA from pooled samples representing several adult structures. The results of our study surprised us, and caused us to conclude that homologous developmental fields can yield staggeringly different adult forms and that highly similar adult characteristics can arise from non-homologous developmental fields. Moreover, homologous genes may underlie some of the physiologically comparable functions of these analogous adult regions. Although homologous sets of genes are usually expressed in cells that develop from homologous morphogenetic fields, this may not be the case in the pallium. We believe there may be different sets of plausible pallial homologies at two (or more) distinct levels of cellular organization: the phylogenetically continuous developmental lineage, and the key underlying genetic programs that distinguish one cell type from another. We argue that the latter level of homology is plausible but cannot be established until the underlying cause of this similar expression is understood.