Supplementary data related to dissertation work titled "Developmental genomics of catfish skin: Evolution of bony armor and denticles through redeployment of ancient gene regulatory toolkit"

The evolution of skin appendages in vertebrates reflects a balance between deep developmental conservation and repeated lineage-specific innovation. An unresolved question is how complex mineralized structures, once lost, can reappear in novel forms. Armored catfishes (Siluriformes) provide a powerful system to address this question, as ancestral scales have been lost and replaced by bony scutes that are often adorned with tooth-like denticles. Here, we investigate the developmental and regulatory basis of scute and denticle formation in the armored catfish Corydoras fulleri using an integrated multi-omic approach combining bulk RNA sequencing, single-nucleus transcriptomics, and chromatin profiling (CUT&Tag/CUTAC).

Bulk RNA-seq differential expression analysis identified genes upregulated during scute and denticle development that are strongly enriched for extracellular matrix components and mineralized tissue regulators, including members of the secretory calcium-binding phosphoprotein (SCPP) gene family, collagens, and key transcription factors, such as sp7 and satb2. In situ hybridization confirmed expression of several of these genes, with osteogenic markers (e.g. bglap or bone gamma-carboxyglutamate protein paralogs) localized to scutes and odontogenic genes (e.g. the SCPP genes enam and scpp9) restricted to denticles. Gene ontology analysis further supported enrichment for processes associated with skeletal development, biomineralization, and odontogenesis. Comparative analyses revealed that while the majority of the differentially-expressed genes are shared with bone, tooth, and scale transcriptomes reported in other vertebrates, there are differences in gene expression between scutes and the scales they replaced.

Single-nucleus RNA-seq provided additional information on the cellular architecture underlying these transcriptional programs, identifying 29 distinct cell populations within the developing skin of C. fulleri. These included a conserved epidermal differentiation continuum, a dermal population containing osteoblast-like cells associated with scute formation, and epithelial clusters linked to denticle development. Notably, denticle-associated populations expressed pitx2and SCPP genes and included an enamel knot–like signaling center marked by dkk1b, bmp2b, msx2b, and fgf20aexpression, indicating deployment of a conserved odontogenic program in the skin. Transcriptional coupling between basal epidermis and dermis further supports conserved epidermal–dermal interactions underlying appendage development.

Chromatin profiling across three teleost species representing scaled (Danio rerio), scute-bearing (C. fulleri), and “naked” (Ictalurus punctatus) skin conditions revealed that regulatory landscapes mirror this combination of conservation and innovation. Motif analysis identified a large, shared enhancer grammar associated with epidermal, pigment cell, and sensory organ developmental programs, alongside a subset of motifs linked to osteogenic pathways shared specifically between zebrafish and armored catfish. Locus-specific analyses demonstrated both conserved and lineage-specific cis-regulatory architectures, including enhancer turnover and reorganization at key genes such as col10a1a, dlx2b, and sp7. Functional reporter assays showed that enhancer competence can be retained even when activity is not deployed in the native developmental context, supporting a model of regulatory flexibility.

Together, these results support a unified model in which the evolution of armored catfish scutes and denticles arises through the redeployment of a conserved vertebrate developmental and regulatory toolkit, combined with lineage-specific modifications in gene expression, localized gene duplication, cell type composition, and enhancer architecture. This work demonstrates that the reappearance of complex mineralized skin appendages does not require the re-evolution of entirely novel programs, but instead reflects the selective reuse, evolvability and reorganization of ancestral vertebrate skin genetic and regulatory mechanisms.