Folding Keratin Gene Clusters During Skin Specification

Regional specification is critical for skin development, regeneration and evolution. The contribution of epigenetics in this process remains unknown. Here using avian epidermis we find two major strategies regulate ß-keratin gene clusters. 1) Over the body, macro-regional specificities (scales, feathers, claws, etc) established by typical enhancers control five sub-clusters located within the epidermal differentiation complex on chromosome 25; 2) Within a feather, micro-regional specificities are orchestrated by temporo-spatial chromatin looping of the feather ß-keratin gene cluster on chromosome 27. Analyses suggest a 3-factor model for regional specification: competence factors (e.g., AP1) make chromatin accessible, regional specifiers (e.g., Zic1) target specific genome regions, and chromatin organizers (e.g., CTCF, SATB2) establish looping configurations. Gene perturbations disrupt morphogenesis and histo-differentiation. This chicken skin paradigm advances our understanding of how regulation of big gene clusters can set up a two-dimensional body surface map. Overall design: To understand the transcriptional and epigenetic control of Keratin gene clusters during embryonic skin specification, we examined region-specific expression patterns and quantified both gene expression and histone modifications of major Keratin clusters at three stages of skin development. We used embryonic day 7 (E7) dorsal back epidermis, which shows no region-specific development, E9 dorsal back and leg epidermis, which shows the beginning of feather specification on the back but no specification on legs yet, as well as E14 dorsal back and leg epidermis, which shows completely specified feather and scale epidermis, respectively. To test the hypothesis that intra-cluster chromatin looping controls differential Feather Keratin (FK) gene expression of the Chr27 ß-keratin cluster, we examined whether higher-order chromatin interactions occur within the cluster in chick embryonic skin epidermis using Next-Generation Capture-C. We used dorsal skin epidermis (experimental group) and whole brain (control group) collected at different developmental stages (E7 and E14). We selected six sites, Pk1, Pk12, Pk18, Pk28, and FK42, randomly across the whole Chr27 ß-keratin gene cluster to design Capture-C probes/baits. We generated two duplicate Capture-C libraries for each tissue. Since skin is afforded additional functions by the presence of diverse skin appendages in different regions, we classified “regional differences” into macro and micro differences, indicating differences across different body parts (feather versus scale) and within an appendage (dorsal and wing feather barb branch), respectively.