ESI Systems Neuroscience Conference 2022

The ever changing brain: Through development and evolution

Comparative neuroanatomists in recent years have made great progress in delineating the conserved “Bauplan” of vertebrate brains. However, brains also evolved some major “new” brain regions that do not fit neatly into the conserved Bauplan view. In particular, it appears that ray-finned fishes and amniotes have independently evolved a dorsal pallium, which comprises the neocortex in mammals. Even for homologous brain areas, changes in connectivity have been common – and often convergent between distantly related species, such as primates and birds. To expand our understanding of brain evolution, consider an analogy to evolving molecular systems. In this analogy, highly conserved genes (including “human disease genes”) correspond to conserved brain regions, whereas genetic losses and gains correspond to the loss and gain of major brain regions. Moreover, molecular pathways and interactomes are analogous to neuroanatomical circuits; they, too, exhibit significant variation, nicely captured by the phrase “genetic rewiring.” Because of this molecular variation, the systems-level function(s) of individual genes and proteins can vary significantly over evolutionary time. This “causal drift” makes it difficult to extrapolate from “simple systems” to more complex networks, and from model species to humans. Analogs of causal drift in molecular networks have been suggested also for neuronal networks (e.g. neocorticalization), but they have not been studied thoroughly. Filling that knowledge gap would be worthwhile.

We study the evolution of animal forms at all scales, with a particular focus on the origin and rise of their most fascinating trait, which is the centralized nervous system. For this, we track the evolution of neurons and other constituent cell types across animal phylogeny, focusing on slow-evolving animals with moderate amounts of secondary loss. We have chosen the nereid Platynereis dumerilii as a powerful model for comparative studies, with morphologically similar organisms already existing as early as the Cambrian. We take advantage of its highly stereotypic development to establish the link between gene expression, cellular morphology, and organ formation for an entire body.
To exemplify the evolution of form in the bilaterian brain, I will explain how we trace the assembly of conserved synaptic proteins, transmitters, and receptors in a basic set of bilaterian neuron types; and how we trace the assembly of these neuron types into a basic set of neural circuits that make up the bilaterian brain. Enabling this, we have constructed a unique cellular atlas for the nereid, the PlatyBrowser, which allows us to combine genome-wide expression profiling with cellular ultrastructure and connectomics for an entire body. To systematically characterize cellular morphologies we have added AI-based recognition of cellular MorphoFeatures to the atlas that we can now align with cell type-specific gene expression modules. I will explain how we use these unique resources to advance our understanding of bilaterian brain evolution and to find hotspots of cellular variation at the micro- and macroevolutionary scale that drive the evolution of form.

Long before brains and nervous systems evolved, organisms of all kinds had to acquire, value, and otherwise process information about existentially salient features of their environment and parameters of their internal functioning. The integrated product of these signaling pathways generated action, including strictly physiological as well as externally observable change. In Principles of Neural Design Sterling and Laughlin (2015) provide the canonical sketch of the chemical basis of cellular information processing in signaling systems found today in unicellular prokaryotes and eukaryotes. Such signaling, internal and external (between and among cells), induces and regulates a wide variety of behavior in single-celled organisms, from solitary and collective motility to nutrient foraging and switching, formation of structured communities (biofilms), and collective, network-based interactions at local and population levels; genomic enhancement via horizontal gene transfer, taking up environmental DNA, and a form of sex; and complex staged developmental sequences (sporulation ) leading to radical change in cell types for long-term survival. Sporulation involves the aggregation of large numbers of cells (tens of thousands plus), but only a proportion survive to germinate another day. Programmed cell death (‘altruistic suicide’) and cellular transformation into specialized support structures claim a substantial proportion of the viable participants. The boldfaced text above reflects known features of brain development and activity. Add to this 1) various types of cell-cell communication, including via secreted signals (packaged in vesicles in some phyla,) and action potentials generated by ion channels (including calcium ion channels); 2) oscillatory activity within single cells as well as in cell collectives; 3) cell migration following chemical gradients (chemotaxis); and 4) large, densely clustered arrays of signaling proteins (several thousands of proteins) capable of detecting and integrating multiple types of stimuli—a highly conserved form of sensorimotor architecture in prokaryotes that has been compared to a ‘nano brain’. In short, the first animals to develop nervous systems and brains already had an information-processing toolkit bristling with options. How these mechanisms are employed to ensure survival, growth, and reproduction in non-neural organisms may hold clues to their recruitment in the origin and evolution of nervous systems and help neuroscientists to better understand brains.

For about a century, bird brains were seen as small, non-cortically organized systems that cast a dim prospect on the cognitive abilities of their beholders. Within the last two decades, this view has changed dramatically. My talk will concentrate on discoveries of about the last decade that demonstrate that birds have a prefrontal-like area with identical functional, electrophysiological, neurochemical, and connectional features as the mammalian prefrontal cortex. Similarly, the avian pallium, although topographically and topologically different from the mammalian one, harbors a connectome akin to those of mammals. In addition, avian neuron numbers are not only much higher than expected by brain size, but also mostly allocated to associative areas in corvids. In parallel, birds developed the ability to cut down the metabolic demands of their neurons by a factor of three. This not only makes a brain with so many neurons affordable but may also provide cellular computational properties that are out of reach for mammals. Lastly, birds even developed a sophisticated cortex within their sensory pallial areas – possibly independent from mammals. Thus, avian and mammalian forebrains converged within 315 million years to an astounding degree. Most importantly, these changes happened very likely in convergent manners without relying on common ancestry. Possibly, evolution does not lack creativity but is just facing a severe limitation of degrees of freedom when wiring a vertebrate brain for sophisticated cognition.

Comparative neuroscience has entered the era of big data. New high-throughput methods for data acquisition (e.g., magnetic resonance imaging, transcriptomics, and sequencing) and data-sharing initiatives have resulted in the availability of large, digital data sets containing many types of data from ever more species at ever more stages of life. The challenge now is to develop a framework for exploiting the new possibilities offered by these data. I will present our ‘common space approach’ for analyzing multi-modal, multi-species data within a single coordinate framework. By comparing brains in terms of abstract feature spaces we can perform quantitative comparisons across very different brains — ranging from humans and non-human primates to rodents — and across different stages of development. This approach allows us to test a new type of hypothesis about how brain organization changes across phylogeny and ontogeny. The approach will also allow us to quantify the limitations of using various non-human animals as ‘model species’ for the human brain and, ultimately, improve their use in translational neuroscience.