In the first-trimester mammalian fetus, skin wounds heal with perfect reconstitution of the dermal architecture without scar formation. Understanding environmental molecular regulation in fetal wound healing may reveal scar-limiting therapeutical strategies for the prevention of postnatal scarring wound repair. Therefore, we performed studies on fetal skin oxygenation and skin and wound expression of hypoxia-inducible factor 1alpha (HIF-1alpha) in the sheep model in vivo and performed studies on the potential relevance of HIF-1alpha during wound healing in vitro. Skin oxygen partial pressure levels were hypoxic throughout normal development. In nonscarring fetal skin at gestation day (GD)60, HIF-1alpha could be detected neither in healthy nor in wounded tissue. At GD100, in wounds with minimal scar formation, HIF-1alpha was expressed in fibroblasts and was markedly up-regulated at the wound edge. In scarring fetal wounds at GD120, HIF-1alpha was predominantly expressed in inflammatory cells. Expression of transforming growth factor beta3 (TGF-beta3), a potent antiscarring cytokine, overlapped with HIF-1a expression at GD100. HIF-1alpha-deficient mouse embryonic fibroblasts showed impaired migratory capabilities and demonstrated that TGF-beta3, but not proscarring TGF-beta1, manifests hypoxia- and HIF-1alpha-dependent regulation. In conclusion, HIF-1alpha-dependent regulation of a potent antiscarring cytokine may provide new strategies for antiscarring manipulation of wound healing.

Fetal skin wounds heal scarlessly while adult wounds scar. Fetal wound healing occurs in a physiologically hypoxic environment whereas in adult wound healing, cells have to acutely adapt to hypoxia caused by locally impaired blood supply. We examined the expression of hypoxia-inducible factor 1 (HIF-1), a potent transcriptional regulator of oxygen-dependent genes such as vascular endothelial growth factor (VEGF), and transforming growth factor-beta (TGF-beta), a potentially HIF-1-regulated scarring cytokine, on fetal and adult responses to wounding. Incisional skin wounds were created in four sheep fetuses (twins served as controls) and two ewes at 100 days of gestation (term = 150 days). Fetal and adult wounds as well as non-wounded control tissues were harvested 2 days post-wounding. Intraoperative arterial blood gas analyses and invasive subcutaneous pO2 measurements revealed that the fetuses were indeed hypoxic while the mothers were normoxic. Expression patterns of HIF-1alpha were investigated by Western blot analyses. HIF-1alpha expression in fetal wounds and fetal control skin was similar, whereas HIF-1alpha was only detected in adult wounds but not in adult control skin. Exposure of cultured fetal and adult dermal fibroblasts to hypoxia (1% O2) showed a marked induction of VEGF mRNA. In contrast, exposure of these cell types to hypoxia did not significantly affect TGF-beta1 mRNA expression in comparison to their normoxic controls. The presence of HIF-1alpha in fetal but not in adult normal skin indicates that HIF-1alpha might be involved in fetal skin development. Conversely, the upregulation of HIF-1alpha in adult but not early fetal wound repair might represent a pathway in the pathogenesis of scarring, since several growth factors overexpressed in, and associated, with scarring are hypoxia-inducible. Further studies need to be performed in order to identify hypoxia-regulated HIF-1alpha target genes involved in the pathogenesis of scarring.

The neuronal circuits mediating the sedative action of diazepam are unknown. Although the motor-depressant action of diazepam is suppressed in ␣1(H101R) homozygous knockin mice expressing diazepam-insensitive ␣1-GABA A receptors, global ␣1-knockout mice show greater motor sedation with diazepam. To clarify this paradox, attributed to compensatory up-regulation of the ␣2 and ␣3 subunits, and to further identify the neuronal circuits supporting diazepam-induced sedation, we generated Emx1-cre-recombinase-mediated conditional mutant mice, selectively lacking the ␣1 subunit (forebrain-specific ␣1 Ϫ/Ϫ ) or expressing either a single wild-type (H) or a single point-mutated (R) ␣1 allele (forebrain-specific ␣1Ϫ/H and ␣1 Ϫ/R mice, respectively) in forebrain glutamatergic neurons. In the rest of the brain, ␣1Ϫ/R mutants are heterozygous ␣1(H101R) mice. Forebrain-specific ␣1 Ϫ/Ϫ mice showed enhanced diazepam-induced motor depression and increased expression of the ␣2 and ␣3 subunits in the neocortex and hippocampus, in comparison with their pseudo-wild-type littermates. Forebrain-specific ␣1 Ϫ/R mice were less sensitive than ␣1Ϫ/H mice to the motor-depressing action of diazepam, but each of these conditional mutants had a similar behavioral response as their corresponding control littermates. Unexpectedly, expression of the ␣1 subunit was reduced in forebrain, notably in ␣1 Ϫ/R mice, and the ␣3 subunit was up-regulated in neocortex, indicating that proper ␣1 subunit expression requires both alleles. In conclusion, conditional manipulation of GABA A receptor ␣1 subunit expression can induce compensatory changes in the affected areas. Specifically, alterations in GABA A receptor expression restricted to forebrain glutamatergic neurons reproduce the behavioral effects seen after a global alteration, thereby implicating these neurons in the motor-sedative effect of diazepam.

Induction of erythropoietin (Epo) expression under hypoxic conditions is mediated by the heterodimeric hypoxia-inducible factor (HIF)-1. Following binding to the 3′ hypoxia-response element (HRE) of the Epo gene, HIF-1 markedly enhances Epo transcription. To facilitate the search for HIF-1 (ant)agonists, a hypoxia-reporter cell line (termed HRCHO5) was constructed containing a stably integrated luciferase gene under the control of triplicated heterologous HREs. Among various agents tested, we identified a class of substances called epolones, which induced HRE-dependent reporter gene activity in HRCHO5 cells. Epolones are fungal products known to induce Epo expression in hepatoma cells. We found that epolones (optimal concentration 4-8 μmol/L) potently induce HIF-1α protein accumulation and nuclear translocation as well as HIF-1 DNA binding and reporter gene transactivation. Interestingly, the activity of a compound related to the fungal epolones, ciclopirox olamine (CPX), was blocked after addition of ferrous iron. This suggests that CPX might interfere with the putative heme oxygen sensor, as has been proposed for the iron chelator deferoxamine mesylate (DFX). However, about 10-fold higher concentrations of DFX (50-100 μmol/L) than CPX were required to maximally induce reporter gene activity in HRCHO5 cells. Moreover, structural, functional, and spectrophotometric data imply a chelator:iron stoichiometry of 1:1 for DFX but 3:1 for CPX. Because the iron concentration in the cell culture medium was determined to be 16 μmol/L, DFX but not CPX function can be explained by complete chelation of medium iron. These results suggest that the lipophilic epolones might induce HIF-1α by intracellular iron chelation.