Homotypic neural circuits of respiratory control – HOMORESP

We will decipher the functional anatomy of the understudied “reticular formation” of the lower brainstem, at cellular resolution, with a focus on respiratory circuits and their necessary coordination with non-respiratory ones. We previously showed that many respiratory neurons fall into two large types, each defined by one developmental transcription factor: i) the pan-autonomic gene Phox2b (for the homeostatic regulation of breathing and the motor command of orofacial muscles —the ancestral respiratory muscles now partially diverted to non-respiratory functions); and ii) the dorsoventral patterning gene Dbx1 (for rhythmic entrainment and the motor command of respiratory pump muscles). Phox2b and Dbx1 neurons engage in many Phox2b>Phox2b and Dbx1>Dbx1synapses, that we call “homotypic”. These observations inspire the tools and questions of the present project:
1) In a first Aim, we will unveil the diversity of neuronal types within the Phox2b and Dbx1 respiratory populations by single cell transcriptomics, on FACS-sorted hindbrains of embryos and newborn animals in which all Phox2b+ neurons or all Dbx1+ neurons are labelled by a fluorescent conditional reporter, and we will cluster neuronal populations by a bioinformatics analysis. In this way, we expect to reveal a rich landscape of hindbrain interneurons. This will be the first effort of that kind, to our knowledge, to address this important part of the central nervous system with this technology;
2) In a second Aim, we will trace the connectivity of Dbx1 and Phox2b neurons (and subset thereof) — including their “homotypic” connections — by a combination of techniques:
i) retrograde monosynaptic glycoprotein-defective rabies viruses, complemented by glycoprotein-encoding helper viruses, either injected into oro-facial muscles, or stereotactically injected in the medulla;
ii) retrograde monosynaptic glycoprotein(G)-defective rabies viruses, G-complemented by a Cre-recombinase-dependent conditional transgene; iii) by a novel, specially designed, transgenic tool that, conditionally and intersectionally, labels the synaptic boutons of specific classes of neurons defined by their expression of two genes.
Using a combination of these techniques we will explore several layers of connectivity within respiratory circuits, in particular those that could explain the successful entanglement, in terrestrial vertebrates, of breathing with non-respiratory oro-facial activities (chewing, swallowing, vocalizing, etc…) that ensure both the “primacy of breathing” and the possibility of transiently suspending breathing: i) map all monosynaptic inputs to the respiratory pacemaker (the preBötzinger Complex), including an hypothesized projection from the recently discovered Post-inspiratory Complex; ii) establish the connectivity matrix of premotor neurons for oro-facial motoneurons in the “Intermediate Reticular formation”, that could underlie coordination of all orofacial activities, iii) discover the hierarchical level above oro-facial premotor neurons, including putative orofacial pacemakers of the “Intermediate Reticular formation” and primary sensory neurons.
3) In a third Aim, we will mine the transcriptomic data obtained in the first Aim by a second round of complex ad hoc bioinformatics analyses, looking for the possibility that the genetic monotony of respiratory neurons (forming homotypic Phox2b or Dbx1 circuits) is causal to the establishment of connectivity. We will implement a variety of techniques (ICA (Independent Component Analyses), NMF (Non-Negative Matrix Factorization) and Bayesian approach to single-cell differential expression analysis) to look for genetic modules that are shared by Phox2b neurons or Dbx1 neurons which are interconnected, and not in neurons which are not. We might thus uncover an original bridge between the world of transcription factors and that of physiologically meaningful neuronal ensembles.