Morphogenesis is transcriptionally coupled to neurogenesis during peripheral olfactory organ development

Abstract

Sense organs acquire their distinctive shapes concomitantly with the differentiation of sensory cells and neurons necessary for their function. While our understanding of the mechanisms controlling morphogenesis and neurogenesis in these structures has grown, how these processes are coordinated remains largely unexplored. Neurogenesis in the zebrafish olfactory epithelium requires the bHLH proneural transcription factor Neurogenin1 (Neurog1). To address whether Neurog1 also controls morphogenesis, we analysed the migratory behaviour of early olfactory neural progenitors in neurog1 mutant embryos. Our results indicate that the oriented movements of these progenitors are disrupted in this context. Morphogenesis is similarly affected by mutations in the chemokine receptor gene, cxcr4b, suggesting it is a potential Neurog1 target gene. We find that Neurog1 directly regulates cxcr4b through an E-boxes cluster located just upstream of the cxcr4b transcription start site. Our results suggest that proneural transcription factors, such as Neurog1, directly couple distinct aspects of nervous system development.

Intertissue mechanical interactions shape the olfactory circuit in zebrafish

“…To evaluate the contribution of other surrounding tissues, we also tracked ngn1:gfp + cells in the adjacent forebrain, and skin (peridermal) cells overlying the OP and the eye. Consistent with previous findings 27,29 , skin cells showed spatially limited, non-oriented displacements. Brain ngn1:gfp + cells did not show directed lateral movements but often moved back and forth along the mediolateral axis, akin to interkinetic nuclear migration within the brain neuroepithelium (Figure 2B and Figure S2E).…”

“…This approach thus provides a temporal control on ECM perturbation, but no precise spatial control. To specifically target the lateral movements in the OP, and not the anteroposterior convergence movements, we performed the injection at 16-18s, when most of the convergence has already occurred 29 . As expected from previous studies reporting a role for the ECM in optic cup morphogenesis 30,42,66 , the collagenases injection resulted in a range of eye defects (from little or no apparent defect to strong invagination phenotypes often associated with extrusion of the lens), likely reflecting variations in the injected volume, a parameter we can not unfortunately finely tune.…”

“…Cxcr4b/Cxcl12a chemotactic signaling has been shown to control OP coalescence downstream of the Neurogenin1 transcription factor 29,35 . While the convergence of the anterior OP cells is particularly affected in mutants for the Cxcr4b/Cxcl12a pathway, the mediolateral component of the cell movements appears to be less perturbed 29 . Conversely, the lateral movements were decreased in eyeless rx3 -/- mutants, but we did not detect significant defects in anteroposterior convergence movements, although the anteroposterior OP dimension was increased at the end of OP coalescence (24s).…”

“…In the embryo, the placodes assemble through the coalescence of progenitors located in large domains of the pan-placodal region 24,25 , while or soon before initiating axonal contacts with the brain 26,27 . During their assembly, neurogenic placodes are surrounded by several non-placodal tissues undergoing morphogenetic reorganisation, including the brain 27,28,29 , the optic cup 30 and neural crest cells 28,31 . The superficial location of the zebrafish neurogenic placodes, right underneath the epidermis, facilitates live imaging of cell dynamics and biomechanical manipulation.…”

“…Remarkably, axon growth initiates during lateral movements through retrograde extension, whereby cell bodies move away from their axon tips attached to the brain 27 (Figure 1A). While the anteroposterior active convergence is guided by Cxcr4b/Cxcl12a signaling downstream of the transcription factor Neurogenin1 29,35 , the mechanisms driving the lateral phase of OP cell movements, during which axons elongate, remain elusive. We previously proposed that the lateral displacement of the cell bodies is a passive, non-autonomous process, based on the following findings: (1) the lack of protrusive activity and of polarised actin and myosin II in the laterally moving cell bodies, (2) the absence of lateral movement defects upon inhibition of microtubule polymerisation or upon cell-autonomous perturbation of actomyosin activity and (3) the lack of lateral movement defects when we disrupted the axons or their attachment to the brain 27 .…”