Reprogramming of keratin biosynthesis by sulforaphane restores skin integrity in epidermolysis bullosa simplex

Abstract

Epidermolysis bullosa simplex (EBS) is a rare inherited condition in which the epidermis loses its integrity after mechanical trauma. EBS is typified by the dysfunction of intermediate filaments in basal keratinocytes of epidermis. Most cases of EBS are due to mutations in the keratin 5 or 14 gene (K5 and K14), whose products copolymerize to form intermediate filaments in basal keratinocytes. Available treatments for this disorder are only palliative. Here we exploit functional redundancy within the keratin gene family as the basis for therapy. We show that genetic activation of Gli2 or treatment with a pharmacological activator of Nrf2, two transcription factors eliciting distinct transcriptional programs, alleviates the blistering caused by a K14 deficiency in an EBS mouse model, correlating with K17 induction in basal epidermal keratinocytes. Nrf2 induction is brought about by treatment with sulforaphane, a natural product. Sulforaphane thus represents an attractive option for the prevention of skin blistering associated with K14 mutations in EBS.

Fig. 1.

Gli2-induced K17 expression rescues K14^−/− mice. (A) Severe blistering on the forepaw of a P2.5 K14^−/− mouse, which is dramatically reduced in a littermate K14^−/−/Gli2^TG mouse. (B) Comparing epidermal histology and biochemical markers in K14^−/− and K14^−/−/Gli2^TG mouse tissue at P2.5. H&E-stained sections of forepaw skin show blistering (bl) in K14^−/− mice but not in K14^−/−/Gli2^TG mice. There is a marked up-regulation of K17 in the epidermis of K14^−/−/Gli2^TG mice. bl, blister; epi, epidermis; hf, hair follicle. (Scale bars: 100 μm.)

Fig. 2.

SF induces K16 and K17 in skin keratinocytes. (A and B) Cultures of skin keratinocytes were treated with 1 μM SF in acetonitrile vehicle. (A) Relative levels of key mRNAs at 12 and 24 h after treatment, relative to acetonitrile-treated controls, and normalized to baseline. GAPDH serves as internal control. Results (mean ± SEM) are from three experiments. Representative results are shown in the upper right corner. (B) Immunostaining for K17 and K14 in acetonitrile-treated (Cont) and SF-treated cells 24 h after treatment. (C and D) Adult SKH-1 hairless mice were topically treated on their backs with 1 μmol of SF in jojoba oil (100 μl) or jojoba oil alone (Cont) for 4 weeks. Treated tissue was harvested 5 days after the last application and processed for morphological analyses (C) or protein extraction, electrophoresis, and Western blotting (D). In D, levels of K17, K16, and K14 antigens are assessed relative to actin. epi, epidermis; hf, hair follicle. (Scale bars: 100 μm.)

Fig. 3.

SF treatment of pregnant K14^−/− mothers causes induction of K17 in fetal epidermis. Tissue from newborn mice was harvested 24 h after the mothers received the last of three injections of 5 μmol of SF or vehicle, or were untreated, and evaluated by protein extraction and analysis (A), indirect immunofluorescence (B), or H&E staining (C). In A, tubulin is the loading control. In B, arrows denote induced K17 in SF-exposed tissues. In C, blistering (bl) is seen in control K14^−/− null pups but not in K14^−/− pups born from SF-treated mothers. The tissue source is indicated at left. epi, epidermis; hf, hair follicle. (Scale bars: 100 μm.)

Fig. 4.

SF treatment reduces skin blistering in newborn K14 null mice. (A–F and A′–F′) Pups receiving either no treatment (Cont) or SF shown at P0.5, P2.5, or P4.5. At P4.5, a wild-type (WT) mouse is shown for comparison (E and E′). (A′–F′) Detailed views of forepaw, an area prone to blistering. (G–J) Forepaws were surgically removed from P2 mice and processed for embedding, sectioning, and H&E staining. Micrographs are from digits in G and I and from dorsal forepaw skin in H and J. Boxed areas in G and I are detailed in G′ and I′. bl, blistering; epi, epidermis; hf, hair follicle; n, nail. (Scale bars: 100 μm.)