Back

HDAC4/5 regulate epidermal barrier function by modulating the epigenetic landscape of human keratinocytes

Nguyen Van, C.; Denis, S.; Cadau, S.; Pelletier, N.; Andre, V.; Lamartine, J.

2026-08-18 cell biology
10.64898/2026.08.13.741670 bioRxiv
Show abstract

Keratinocyte proliferation and differentiation are essential to produce the stratified structure of the epidermis and maintain its barrier function. These processes are regulated by complex mechanisms including epigenetic regulation. In this study, we evaluated the role of HDAC4/5, two class IIa histone deacetylases, in the epigenetic regulation of proliferative and differentiated human keratinocytes using dedicated 2D and 3D in vitro models. Our findings demonstrate that chemical inhibition or shRNA-mediated knock-down of HDAC4 impair keratinocyte proliferation notably through increased H3K27 acetylation and subsequent transcriptional activation of the cell cycle inhibitor gene BTG2. Interestingly, HDAC4/5 inhibition alters H3K27 acetylation landscape in proliferating keratinocytes, whereas the epigenetic identity of differentiated keratinocytes is much less affected. Inhibiting HDAC4/5 in 3D epidermis models resulted in reduced epidermal thickness and impaired barrier function linked to alteration in the lipid composition of the stratum corneum. Furthermore, analysis of several well-established skin aging markers revealed that reconstructed human epidermis treated with the HDAC4/5 inhibitor exhibit molecular and functional characteristics consistent with an aged-epidermis. Collectively, our results demonstrate that HDAC4/5 are essential for maintaining epidermal homeostasis and pave the way for the development of innovative models of skin aging based on the modulation of histone acetylation.

Matching journals

The top 9 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.