Back

Transient activation of potent progenitor cells is required for spinal cord regeneration

Weinholtz, C. A.; Zhou, L.; Saraswathy, V.; Xu, Y.; Shaw, D. K.; McAdow, A. R.; Park, D.; Shin, J.; Solnica-Krezel, L.; Johnson, A. N.; Mokalled, M.

2026-02-07 neuroscience
10.64898/2026.02.04.703854 bioRxiv
Show abstract

Adult zebrafish exhibit full recovery following spinal cord injury. Transient expansion of stem cell-like progenitors is thought to underlie their regenerative capacity. Yet, our understanding of the identities and contributions of the crucial stem cell populations that direct spontaneous neural repair remains limited. Moreover, while most neural regeneration research is centered on promoting proliferative repair, the regulatory mechanisms that reinstate quiescence post-repair are unknown. Here, we determined the molecular identities and cellular contributions of sox2+ progenitors during spinal cord repair. Genetic lineage tracing shows zebrafish spinal progenitors, while quiescent in uninjured tissues, self-renew and differentiate into neurons and glia after injury. By single-cell sequencing, sox2+ cells are heterogeneous and biased towards neuronal or glial fates in both homeostatic and regenerating tissues. By screening for transcription factors that are differentially expressed in acute versus chronic spinal cord injury, we find the Bach1 transcription factors control transient progenitor cell activation by acting as dual activators and repressors of sox2 expression. This study elucidates the molecular diversity and contributions of sox2 expressing cells during spinal cord repair and identifies a transcriptional regulatory switch by which progenitor cells expand after injury and restore quiescence after regeneration is completed.

Matching journals

The top 4 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.