Back

A Hedgehog-Induced Homeodomain Cascade Couples Progenitor Identity to Neuronal Diversification

Xu, C.; Sastry, R.; Newstein, P.; Delgado, R.; Wagh, N.; Silies, M.; Cepko, C. L.; Doe, C.

2025-09-07 developmental biology
10.1101/2024.09.27.615552 bioRxiv
Show abstract

The brain deploys diverse neuronal subtypes to split complex inputs into parallel channels--each tuned to distinct features--enabling rich neural processing. Yet how progenitors generate distinct but functionally related subtypes remains unknown. In the Drosophila lamina (five lamina neuron subtypes receiving photoreceptor input), we uncover the regulatory logic: a pan-class homeodomain transcription factor (HDTF), induced by Hedgehog in progenitors and maintained in all lamina neurons, drives diversification within the lamina neuron class by orchestrating a four-step program across the progenitor-to-newborn neuron transition. Specifically, it establishes progenitor identity, promotes cell-cycle exit, induces subtype-specific HDTFs, and acts as their obligate cofactor to specify distinct subtypes. Loss of subtype-specific HDTFs in newborn--but not older--neurons drives subtype-to-subtype fate conversions at molecular, morphological, and functional levels, including a contrast-to-luminance encoding switch. In the mouse retina, we find that each of the 63 amacrine, 15 bipolar, and 45 retinal ganglion cell subtypes expresses pan-class and subtype-specific HDTFs, indicating evolutionary conservation of this regulatory logic. Given the brain-wide expression of HDTFs across species, these findings convert a longstanding mystery into a testable, generalizable principle for within-class subtype diversification and lay the groundwork for subtype-precise reprogramming and cell replacement strategies.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.