‘Don't be so over‐protective!’

Erwin Tschachler, Leopold Eckhart, Florian Gruber

Author Affiliations

  • Erwin Tschachler, 1Department of Dermatology, Medical University of Vienna, Vienna, Austria
  • Leopold Eckhart, 1Department of Dermatology, Medical University of Vienna, Vienna, Austria
  • Florian Gruber, 1Department of Dermatology, Medical University of Vienna, Vienna, Austria

See related article in EMBO Molecular Medicine

Oxidative stress within tissues and organs results from a failure to detoxify reactive oxygen species (ROS) and leads to damage of cellular components including proteins, DNA and lipids. This damage is thought to be a key mechanism of aging as well as to be involved in the pathogenesis of so diverse diseases as diabetes, schizophrenia, arteriosclerosis and many more (Ghosh et al, 2011; Negre‐Salvayre et al, 2010). It is therefore not surprising that the search for strategies to thwart oxidative stress has been at the centre of research efforts since several decades. Anti‐oxidants have been tested as nutrient additives or, as far as the skin is concerned, in topical formulations with variable success. Apart from providing exogenous anti‐oxidants, boosting the organism's own anti‐oxidant defense has been proposed as a promising way forward.

The redox‐regulated transcription factor ‘Nuclear factor, erythroid derived 2, like 2’ (Nrf2) is a central switch in the cellular defense against oxidative stress and agonists of Nrf2 are currently tested in clinical trials (Pergola et al, 2011). Under baseline conditions, Nrf2 is retained in the cytoplasm and proteolytically degraded through binding to Keap1. On exposure to electrophils, Nrf2 is released from Keap1 and re‐localizes to the nucleus where it promotes transcription of a wide range of cytoprotective enzymes and cellular anti‐oxidants (Osburn & Kensler 2008). Inactivation of the Nrf2 gene is compatible with life (Chan et al, 1996) but leads a dramatic reduction of the resistance to oxidative stress (Li et al, 2004). Inactivation of Keap1 results in constitutive activation of Nrf2 and upregulation of anti‐oxidant and detoxification genes, but surprisingly also leads to aberrant differentiation of oesophageal keratinocytes and death from malnutrition after weaning (Wakabayashi et al, 2003). Another mouse model expressing a transgene of a constitutively active Nrf2 in epidermal keratinocytes developed normally and showed a higher level of protection against UV‐induced cellular damage (Schäfer et al, 2010).

In their work published in this issue of EMBO Molecular Medicine, Schäfer and colleagues (Schäfer et al, 2012) used a modified transgene with a fivefold higher expression level of the constitutively active Nrf2 than in the previously reported model. In contrast to the Keap1 knockout animals, no oesophageal pathology was observed and the animals survived into adulthood. However, the transgenic animals developed a pronounced skin phenotype likened by the authors to certain forms of ichthyosis, with a thickened epidermis, a thickened stratum corneum and the presence of inflammatory cells. The skin changes, which developed in the weeks after birth, were ascribed to a defective skin barrier manifesting with increased transepidermal water loss, increased fragility of the corneocytes and alterations of the stratum corneum lipids. Small proline rich proteins 2d and 2h (Sprr2d and 2h) and the secretory leukocyte peptidase inhibitor (Slpi) showed a significant regulation already before the onset of hyperkeratosis and skin inflammation. These new epidermal Nrf2 targets were previously shown to act as ROS scavengers and anti‐microbial, respectively, suggesting that they were typical stress response factors induced by Nrf2. In addition, they are known modifiers of the mechanical resilience of corneocytes and of the proteolytic desquamation of corneocytes, providing a likely explanation for the primary defect in Nrf2 transgenic mice. This hypothesis was corroborated by experiments in which Nrf2 agonists were applied in high amounts onto the skin of mice. Like the introduction of the Nrf2 transgene, this led to an upregulation of both Sprr2 and Slpi and to an increased epidermal thickness, which could be blunted by deletion of the endogenous Nrf2 gene.

Among the skin abnormalities, which appeared after initial alterations of cornification and corneocyte shedding, the disturbance of the lipid barrier is particularly intriguing. Previously, Nrf2 was shown to be activated by oxidized lipids (Gruber et al, 2010) and several Nrf2‐regulated genes involved in lipid metabolism, such as CD36, ELOVLs, phospholipases (PRDXs) and PPARγ, are critical for the formation of the epidermal lipid barrier (Feingold, 2007). Although, Elovls are not up‐regulated before the onset of the skin phenotype in the constitutively active Nrf2 transgenic mice it will be worth analysing what drives the lipid abnormalities observed at later time points in these animals and what the contributions of the above Nrf2 targets to the skin lipid barrier are.

Taken together, the findings reported by Schäfer et al are the first to link Nrf2 action to the modulation of the skin barrier.

. . .the findings reported by Schäfer et al are the first to link Nrf2 action to the modulation of the skin barrier.

The candidate mediators of Nrf2's potentially negative effects identified by Schäfer et al should now be studied for their individual contributions to skin barrier defects. In addition, it is important to note that the strong over‐expression of Nrf2 in all keratin 5‐expressing keratinocytes caused abnormalities of hair follicles and sebaceous glands that were not investigated in the present report (Schäfer et al, 2012). As these alterations are likely to influence the function of the interfollicular epidermis, a better understanding of the phenotype of this new mouse model urgently requires in‐depth characterization of the effect of continuous Nrf2 over‐activation on the pilosebaceous unit.

While in the past the effects of Nrf2 activation have been generally found to be beneficial for various tissues under stress conditions, a remarkable example of Nrf2‐mediated tissue damage has been reported previously: the genetic suppression of protein degradation by autophagy in the liver led to the activation of Nrf2 and to the accumulation of Nrf2‐induced proteins such as glutathione S‐transferase‐m1 which caused severe liver injury (Komatsu et al, 2010). The report of Schäfer et al extends this concept and suggests that over‐activation of Nrf2 does not protect but disrupt epidermal homeostasis.

The authors declare that they have no conflict of interest.