PRDM3 and PRDM16 define cranial neural crest cell states in zebrafish development

Background

Cranial neural crest cells are multipotent embryonic cells that form many structures in the head, including cartilage, bone, pigment cells, neurons and glia. Their differentiation requires precise control of gene expression and chromatin accessibility. PRDM3 and PRDM16 are related chromatin-associated proteins that are important for craniofacial development. Previous studies showed that losing these proteins disrupts neural crest differentiation, particularly cartilage formation. However, the genomic regions associated with PRDM3 and PRDM16 in developing neural crest cells were not well understood and their direct transcriptional and chromatin targets in neural crest development remain largely unknown.
In this preprint, Shull and colleagues investigated which chromatin regions are associated with PRDM3 and PRDM16 in zebrafish neural crest cells at 48 hours post-fertilization, a stage when many cranial neural crest cells have completed migration and are beginning to differentiate into specialized cell types.

Key findings

The authors isolated sox10mRFP-positive neural crest cells using FACS and performed CUT&RUN for PRDM3, PRDM16 and H3K27ac. CUT&RUN is an antibody-guided technique that cuts and releases DNA located near a protein of interest, allowing researchers to identify the chromatin regions associated with that protein. IgG was included as a negative control.
Results show that most PRDM3- and PRDM16-associated DNA fragments were 120-200 base pairs long, like the amount of DNA protected by a nucleosome. The authors therefore proposed that both proteins are mainly associated with nucleosome-containing chromatin rather than directly binding exposed DNA at this stage. However, these fragment sizes show nucleosome association or proximity and do not yet prove direct physical binding to nucleosomes. The authors combined PRDM3, PRDM16 and H3K27ac peaks and separated them into six clusters. Some regions were mainly associated with PRDM3, some with PRDM16 and others with both proteins. Genes located near these regions were linked with neural crest migration, Wnt signalling, neuronal development, pigmentation and craniofacial differentiation.
Comparison with previous ATAC-seq data suggested that PRDM3 and PRDM16 may have different effects on chromatin. Loss of PRDM3 increased accessibility at some regions, whereas loss of PRDM16 reduced accessibility. This led the authors to suggest that PRDM3 may promote chromatin compaction, while PRDM16 may support chromatin accessibility.
Finally, the authors examined selected candidate genes using insitu hybridization chain reaction (HCR) and CUT&RUN-qPCR. Some genes showed altered expression in PRDM3 or PRDM16 mutants, and PRDM enrichment was reduced at several tested regions in the corresponding mutants. These only results partly matched the proposed opposing roles of PRDM3 and PRDM16 in chromatin accessibility.

Why I chose this preprint

I liked that the study combined chromatin binding, accessibility and spatial gene-expression analysis. The fragment-size analysis was interesting because it suggests that PRDM3 and PRDM16 may act within nucleosome-associated chromatin. The study also indicates that these two related proteins are not completely similar and that they occupy some common regions but also show distinct binding patterns and potentially different effects on chromatin accessibility. This work may be a useful starting point for understanding how PRDM family proteins may influence neural crest differentiation and craniofacial development.

Questions for the authors

1. Which specific neural crest cell types show each PRDM3 and PRDM16 binding pattern? I think maybe single-cell method could help separate cartilage, neuronal, glial and other-cell populations?
2. Do PRDM3 and PRDM16 physically interact with nucleosomes, or are they recruited indirectly through maybe other proteins?
3. Is the methyltransferase activity of PRDM3 or PRDM16 required for these effects? What if you make the proteins catalytically inactive?
4. How do PRDM3 and PRDM16 chromatin interactions change after 48 hpf? Comparing earlier and later stages could reveal when these proteins change or maintain neural crest cell states.
5. What does Cluster 4 really represent? The paper describes it inconsistently, so the authors may clarify whether these regions are active, inactive, or a mixture of different chromatin states.

(No Ratings Yet)