Fig 2: Compromised epidermal integrity in Grhl3–/– skin is improved through application of 5ASA.Back skins from E16.5 WT and Grhl3–/– embryo were cultured together with vehicle control (a, c) or in the presence of 1 mM 5ASA (b, d). No significant changes in WT epidermal morphology were observed following 5ASA treatment of WT skin, however, treatment of the Grhl3–/– skin (c, d) restored granular layer formation. The epidermal differentiation marker, K10, was also specifically increased in Grhl3–/– skins following 5ASA treatment (e-h), and expression of the structural barrier integrity protein PPL (i-l) was restored. 5ASA did not affect the skin from WT animals (m, n) but reduced the expression of TARC in Grhl3–/– skin (o, p). The treatment of E16.5 WT skin with recombinant TARC demonstrates expansion of the basal layer Keratin 14-positive compartment and reduced keratinocyte compaction towards the granular layer (q-v). The granular layer is less pronounced in H&Es (w, x). Scale bars correspond to 100 µm. These data are summarised in the proposed functional model (y) suggesting that loss of GRHL3 in differentiated cells leads to a compromised epidermal barrier with concomitant TARC production and secretion to stimulate proliferation of basal keratinocytes. This triggers additional TARC expression from basal cells amplifying the proliferative signal

Fig 3: Loss of GRHL3 leads to increased keratinocyte cell proliferation and elevated TARC expression.KD of GRHL3 via shRNA (shRNA1-GRHL3) leads to increased colony size (a, b), relative to transduction with empty vector (sh-EV), albeit not increased colony number (data not shown) in the human epidermal keratinocyte cell line HaCaT. Conditioned medium (CM) from EV cultures did not stimulate growth of shRNA1 cells (c). However, CM from shRNA1 cultures stimulated significant growth of EV colonies (d), a finding confirmed by quantitation of total colony area per plate (e). Analysis of cytokine activity in CM collected from HaCaT cells transduced with sh-EV (f) or shRNA1-GRHL3 (g) shows that the only cytokine that is significantly overexpressed (when quantitated by densitometric scanning; h) following GRHL3 KD is TARC (dotted white box). Positive (red boxes) and negative controls (yellow boxes) are also shown. Q-RT-PCR quantitation of mRNA expression (i) of E18.5 back skin from WT and Grhl3–/– embryos shows that TARC is significantly elevated in the skin of Grhl3–/– embryos. Immunohistochemical analysis (j, k) shows predominant TARC expression in the nuclei of keratinocytes near or in the granular layer. Weaker basal TARC is also detected. Scale bars correspond to 50 µm. When WT embryonic skin is placed in a microtube with WT adult blood, (l, m), no clotting occurs, indicative of an absence of soluble clotting factors in the blood. However, when blood from the same adult animal is incubated with skin from Grhl3–/– embryos (n, o), rapid clotting occurs. Ultrastructural analysis of blood on both WT (p) and Grhl3–/– embryo skin (q) by scanning electron microscopy (SEM) shows no clotting of blood on WT skin, but the presence of strands of fibrin (indicative of clotting) is clearly seen on the Grhl3–/– embryo skin. * and ** indicates p < 0.05 and p < 0.01, respectively. Scale bars correspond to 2 µm