Thyroxine Restores Severely Impaired Cutaneous Re-Epithelialisation and Angiogenesis in a Novel Preclinical Assay for Studying Human Skin Wound Healing Under "Pathological" Conditions Ex Vivo

Arch Dermatol Res. 2020 Jun 22. doi: 10.1007/s00403-020-02092-z. Online ahead of print.


Impaired cutaneous wound healing remains a major healthcare challenge. The enormity of this challenge is compounded by the lack of preclinical human skin wound healing models that recapitulate selected key factors underlying impaired healing, namely hypoxia/poor tissue perfusion, oxidative damage, defective innervation, and hyperglycaemia. Since organ-cultured human skin already represents a denervated and impaired perfusion state, we sought to further mimic "pathological" wound healing conditions by culturing experimentally wounded, healthy full-thickness frontotemporal skin from three healthy female subjects for three days in either serum-free supplemented Williams' E medium or in unsupplemented medium under "pathological" conditions (i.e. hypoxia [5% O2], oxidative damage [10 mM H2O2], absence of insulin, excess glucose). Under these "pathological" conditions, dermal-epidermal split formation and dyskeratosis were prominent in organ-cultured human skin, and epidermal reepithelialisation was significantly impaired (p < 0.001), associated with reduced keratinocyte proliferation (p < 0.001), cytokeratin 6 expression (p < 0.001) and increased apoptosis (p < 0.001). Moreover, markers of intracutaneous angiogenesis (CD31 immunoreactivity and the number of of CD31 positive cells and CD31 positive vessel lumina) were significantly reduced. Since we had previously shown that thyroxine promotes wound healing in healthy human skin ex vivo, we tested whether this in principle also occurs under "pathological" wound healing conditions. Indeed, thyroxine administration sufficed to rescue re-epithelialisation (p < 0.001) and promoted both epidermal keratinocyte proliferation (p < 0.01) and angiogenesis in terms of CD31 immunoreactivity and CD31 positive cells under "pathological" conditions (p < 0.001) ex vivo. This demonstrates the utility of this pragmatic short-term ex vivo model, which recapitulates some key parameters of impaired human skin wound healing, for the preclinical identification of promising wound healing promoters.

Keywords: Ex vivo; Hypoxia; Skin model; Wound healing.