Nrf2: the brake in oxidative stress that Nox4 needs to protect the heart

Cardiac muscle subjected to chronic afterload triggers myocyte death, fibrosis, ventricular dilatation and contractile dysfunction which chronically develop as a pathological condition. While many molecular pathways contribute to pathological cardiac hypertrophy and heart failure, one well-established mechanism relates to activation of specific reactive oxygen species (ROS)-generating enzymes and endogenous antioxidants that can both contribute to pathology and induce protective mechanisms to limit overall stress.

NADPH oxidases (Noxes) have been shown to have opposing roles in response to hypertrophic stimuli in the heart. In 2003, Shah and colleagues,¹ showed that Nox2 mediated angiotensin II-induced hypertrophy, but that Nox2 knockout had no effect on pressure-induced hypertrophy. In fact, loss of Nox2 actually increased ROS production, as it turns out because Nox4 is upregulated in this condition. Using models of cardiomyocyte-specific Nox4 overexpression, they went on to show that surprisingly Nox4 was protective against cardiac stress promoted by pressure overload, in part by preserving myocardial capillary density through activation of HIF-1α and the release of VEGF.² Later work by this group demonstrated that cardiomyocyte-targeted Nox4 overexpression activated the transcription factor Nrf2 and consequently led to greater transcription of genes involved in antioxidant responses, such as heme oxygenase, glutathione reductase and glutamatecysteine ligase.³ Collectively, these studies led to the concept that Nox4 is a constitutive and protective NADPH oxidase homologue.

In this issue, Shah and colleagues extend these observations to show that Nox4-mediated protection against cell death and tissue damage after aortic constriction is in fact a consequence of Nrf2 upregulation (Figure 1). Wild-type mice subjected to aortic banding are protected from mitochondrial DNA lesions, present less cell death and have decreased myocardial stress when compared to their Nrf2^−/− counterparts. Unexpectedly, cell death in this model is triggered by a cytochrome c-independent mechanism that activates the AIF/caspase 8/tBid pathway and triggers apoptosis-related outer mitochondrial membrane events. Nrf2 activation was shown to be dependent upon Nox4, because Nox4 knockout mice were unable to upregulate Nrf2 after aortic banding, and mice with cardiomyocyte-specific transgenic overexpression of Nox4 and deletion of Nrf2 (Nox4TG/Nrrf2^−/−) developed less cardiac dysfunction, dilatation and hypertrophy than wild type mice. These data define a novel endogenous link between Nox4 and Nrf2 and suggest that the Nox4/Nrf2 axis may serve as a unique redox rheostat to respond to oxidative stress.

Figure 1. The protective role of Nox4 and Nrf2 in cardiac dysfunction.

Model based on a combination of in vitro and in vivo studies. Pressure overload or phenylephrine-induced cardiomyocyte contraction promotes Nox4 activation. Nox4-driven ROS production is regulated by the recruitment/activation of Nrf2 that triggers transcription of many antioxidant genes, which serves to inhibit DNA damage, oxidative stress and caspase 8-mediated apoptosis. ATF= Apoptosis-inducing factor; AO genes= antioxidant genes; Hmox1= heme oxygenase (decycling) 1; Gclc= glutamate-cysteine ligase, catalytic subunit; Gsta2= glutathione S-transferase alpha 2, tBid=truncated p15 BH3-interacting domain death agonist.

However, there are many scenarios where Nox4 is not able to promote an increase in Nrf2 expression/activity, thus generating a redox imbalance. It has been reported that when mice age, the reparative response to lung injury is impaired, causing a Nox4-mediated persistent fibrosis that is characterized by decrease in Nrf2 expression.⁴ Also, it has been recently shown that doxorubicin nephrotoxicity is marked by increase in Nox4 but no change in Nrf2 expression, and that attenuation by thymoquinone, a potent anti-oxidative and anti-inflammatory drug, is able to revert the kidney toxicity.⁵ These data are consistent with the notion that Nox4 is able to play a protective role only when succeeded by an antioxidant response such as Nrf2 activation, and help to explain pro-pathology functions ascribed to Nox4.

The precise mechanism by which Nox4 regulates Nrf2 and therefore antioxidant gene expression has not been confirmed. The abundance of Nrf2 and its inhibitor KEAP1 in the cell is maintained by a feedback autoregulatory loop in which Nrf2 induces transcription of a Cul3/Rbx1 complex that ubiquitinates and targets Nrf2 and/or its inhibitor KEAP1 to degradation by the proteasome. However, what primarily triggers Nrf2 activation is not clear. Post-translational modifications such as phosphorylation and acetylation regulate its translocation to the nucleus and activation of many target genes (reviewed in ⁶). Common upstream stimulators of Nrf2 expression belong to antioxidant, tocopherol and phytochemical classes and activate Nrf2 by PKC-δ-dependent phosphorylation of its serine 40.⁷ To turn off Nrf2, Fyn, which is phosphorylated by GSK3β, translocates to the nucleus and phosphorylates Nrf2 on Y568, resulting in its nuclear export and degradation.⁸ One likely possibility is that H₂O₂ derived from Nox4 oxidizes KEAP1, targeting it to degradation and releasing Nrf2. For this to be true, we must understand how Nox4, which is expressed in the endoplasmic reticulum in most cells, can affect the cytoplasmic protein KEAP1. Alternatively, Nox4 could regulate Nrf2 phosphorylation, although there are currently no studies linking Nox4 to Nrf2 or KEAP1 kinases. A third possibility is that Nox4 regulates nuclear import or export of Nrf2, a possibility that should be examined since Nox4 has been found in the nucleus in some cells.⁹ Further work is necessary to fully understand how Nox4 modulate Nrf2 activity.

Because Nrf2 has been described as a master regulator of antioxidant responses and defensive genes in many diseases, including neurodegeneration, cancer, kidney disease, cardiovascular diseases, hepatitis, and inflammation associated with infection, the Nox4/Nrf2 pathway may represent a common protective mechanism and therefore an attractive therapeutic target. The universal failure of the antioxidant vitamin trials¹⁰ raised questions about the oxidative theory of cardiovascular disease, but the present work provides additional insight into why treatment with nonspecific antioxidants might not be beneficial. Antioxidant vitamins would tend to scavenge Nox4-induced production of ROS, paradoxically therefore inhibiting Nrf2 activation and the induction of the body's own antioxidant defenses. Important homeostatic cell functions regulated by Nrf2 include glutathione and thioredoxin production, lipogenesis as well as regeneration of NADPH and purine biosynthesis,⁶ suggesting that the Nox4/Nrf2 pathway may be critical for metabolic homeostasis. Several Nrf2 activators have entered clinical trials, and one, dimethyl fumarate, is approved for use in multiple sclerosis. Sulforaphane is currently being tested in clinical trials for prostate cancer, asthma, autism and schizophrenia. Care must be taken, however, when applying this concept to cardiovascular diseases, since Nrf2 can elevate plasma cholesterol and worsen atherosclerosis.¹¹

The intriguing article by Smyrnias et al. in this issue thus provides a strong rationale for further basic, translational and clinical studies focused on the Nox4//Nrf2 axis. It will be important to test the role of this pathway in specific tissues and in specific disease states. Give the ubiquitous nature of Nox4 and Nrf2, future therapies targeted to specific tissues while avoiding systemic effects of activators of these pathways are the most likely to be successful.