Institut de Biologie de l'École Normale Supérieure ENS . CNRS UMR8197 . Inserm U1024 Directeur : Pierre Paoletti

Development and evolution of neural circuits

A long line of biologists, from the physiologist Xavier Bichat (1771-1802) to the paleontologist Alfred Romer (1894 -1973), have described the vertebrate body in general, and its nervous system in particular, as made of two parts, one somatic and one visceral, “imperfectly welded onto each other”: the somatic part deals with “external affairs” (i.e. our relation to the environment), the visceral part deals with “internal affairs” (i.e. homeostasis, through the control of respiratory, cardiovascular and digestive functions). In a striking echo of this view, we discovered some years ago that the majority of the constituent neurons of the visceral circuits (sensory neurons of epibranchial ganglia, ganglionic autonomic neurons, their preganglionic neurons, and several classes of interneurons) share a dedicated transcription factor, Phox2b, a veritable “master gene” of the visceral circuits, which sets them apart from somatic neurons. This ontogenetic unity—which makes visceral reflex circuits, collectively, one organ, and visceral neurons, collectively, one class (or super-class) of neurons—, informs much of our work. We study simultaneously several parts of the visceral circuits, and from several perspectives: embryological, physiological and evolutionary, each line of research contextualized by the others.

From an embryological standpoint, we study the formation of autonomic ganglia (parasympathetic, enteric, pelvic) and of epibranchial (sensory) ganglia. We have shown that parasympathetic ganglia form by aggregation of Schwann cell precursors that migrate along cranial nerves towards the site of ganglion formation. We have shown that, in sensory neurons, Phox2b serves as a switch from a somatic identity (e.g. touch receptors) to a visceral one (e.g. baroreceptors or chemoreceptors). We have shown that the pelvic ganglion, supposedly and mysteriously “mixed” (parasympathetic/sympathetic) according to a century-old dogma, is in fact sympathetic, and that the pelvic organs are exclusively innervated by sympathetic neurons.

From an evolutionary standpoint, we have shown that the central nervous system of adult tunicates (Ciona intestinalis) — essentially «visceral» creatures that spend their life fixed on the sea floor, breathing and feeding — harbors homologues of the vertebrate cranial “branchiomotor” neurons, which are the respiratory motoneurons in fish (thus, cranial motoneurons predate craniates). We have more recently shown that the viscero-somatic duality of sensorimotor circuits is present in mollusks (and thus, predates bilaterians). We are currently searching for respiratory centers in fish that could be compared to those of terrestrial vertebrates.

From a physiopathological standpoint, in collaboration wih the laboratory of Gilles Fortin we have used intersectional genetics in mouse to discover the locus of the central chemoreflex (i.e. hyperventilation in response to high CO2), the “retrotrapezoid nucleus”, a group of interneurons which depend on Phox2b and are the likely culprit in the deadly respiratory symptoms of Congenital Central Hypoventilation Syndrome (CCHS), a human disease caused by mutations in PHOX2B. We have found that Phox2b+ neurons of the reticular formation of the hindbrain were not only premotor to motoneurons of the face and neck (themselves Phox2b+) but control lapping, the rhythmic behavior for ingesting liquids in many terrestrial vertebrates.

Current physiological work, under the direction of Gilles Fortin who joined us at the ENS, concerns respiration, one of the three cardinal visceral functions, with blood circulation and digestion. Respiration is a vital behavior, relatively simple and conserved among vertebrates. In this model, we aim to identify the principles by which neural circuits orchestrate the precise and timely control of behavior. We seek to understand how the nervous system can generate/modulate ongoing rhythmic activity and compute the neural commands that mediate homeostatic aspects of breathing (gas exchange in the lungs) and non-homeostatic ones (e.g., vocalization).
To decipher how brainstem respiratory circuits engage in the control of various functional tasks, we need to unravel the neural subpopulations organized into specific circuits with dedicated executive functions. These questions are addressed through multiple approaches that include state-of-the-art genetics in mice, viral tracing strategies, functional manipulation (opto-/chemo-genetics) and biophysical (e.g., electrophysiology, calcium imaging...) and video tracking for quantitative behavioral analysis. Thanks to investigations at different stages of development, these approaches also touch upon the mechanisms of circuit assembly.