Nuclear factor erythroid 2 related factor 2 (NRF2) plays a central role in protecting cells from oxidative injury. As a transcription factor, NRF2 induces gene expression of enzymes that combat the effects of oxidative stress [36]. NRF2 target genes share a common DNA sequence known as the antioxidant response element (ARE) in their promoter regions that is required for NRF2 binding and gene induction [37]. NRF2 is constitutively expressed in the cytoplasm of all cell types; however, it is normally kept at low levels via Kelch-like ECH-associated protein 1 (KEAP1)-mediated ubiquitination and degradation. This also serves to keep NRF2 target antioxidant genes at the low, basal levels needed to maintain their “housekeeping” functions [38]. When modified during periods of oxidative stress, interaction between reactive oxygen species and cysteine residues of KEAP1 allows NRF2 to escape KEAP1-mediated ubiquitination and degradation and translocate to the nucleus where it can induce ARE-containing targets for the purpose of restoring oxidative homeostasis in the cell [38]. The myriad target genes include antioxidant proteins, phase I oxidation, reduction, and hydrolysis genes, phase II detoxifying enzymes such as glutathione s-transferases (GSTs), NADPH-generating enzymes, drug transporters, and stress proteins involved in heme and metal metabolism, such as heme oxygenase 1 (HO-1) [26].

The role of NRF2 has been studied in both animal models and human CKD. In animal models, under normal, healthy conditions, NRF2 knockout mice exhibit no abnormalities throughout their lifespan. However, in disease state models, such as cardiac disease, diabetes, and obesity, loss of NRF2 augments disease severity [36]. In the kidney, studies using both genetic and pharmacologic approaches have revealed the protective effect of NRF2 in animal models of CKD.

Jiang et al. studied the role of NRF2 in a streptozotocin (STZ)-induced diabetic nephropathy model in Nrf2 wild-type (WT) and knockout (KO) mice [39]. After 16 weeks, despite a similarly achieved level of hyperglycemia, Nrf2 KO mice had a higher degree of oxidative damage, more albuminuria, and more severe glomerulosclerosis, compared to WT mice [39]. In other models of kidney disease, including autoimmune nephritis, toxic injury, ischemia reperfusion injury (IRI), ureteral obstruction, and podocyte injury, Nrf2 KO organisms also displayed an increase in disease severity, suggesting that NRF2 plays a nephroprotective role through a common pathway [39,40,41,42,43,44,45,46,47,48,49,50]. Similar nephroprotection is seen with NRF2 activators and KEAP1 suppressors (which allows NRF2 to translocate to the nucleus and exert its effect) [36]. In mouse models of ischemia and unilateral ureteral obstruction, KEAP1 hypomorphic mice displayed attenuated kidney disease compared to KEAP1-intact mice [47]. Zheng et al. studied the role of the NRF2 activators sulforaphane (more below) and cinnamic aldehyde in a STZ-induced diabetic nephropathy model in Nrf2 wild-type and knockout mice. They found that the NRF2 activators attenuated markers of kidney damage and minimized glomerular pathology in wild-type but not in Nrf2 knockout mice [50]. However, Nrf2 deletion has been shown to be beneficial in a model of autoimmune nephritis, by increasing sensitivity to tumor necrosis factor-alpha (TNF-α)-mediated apoptosis [51]. Similarly, in the Akita mouse model of Type 1 diabetes, genetic deletion or pharmacological inhibition of Nrf2 attenuated hypertension and kidney disease [52]. It is possible that effect of NRF2 is disease context-dependent. Nevertheless, taken together, the data suggest that NRF2 has a nephroprotective role in kidney disease.

In humans, the most well-studied pharmacologic agent activating the NRF2 system is the drug bardoxolone methyl, which covalently binds to cysteine residues of KEAP1, allowing NRF2 to escape ubiquitination and degradation and translocate to the nucleus to induce the myriad antioxidant genes [53]. The potential beneficial effect of bardoxolone methyl on kidney function was first observed in a phase I cancer trial, where it was found to result in a statistically significant increase in estimated glomerular filtration rate (eGFR) of 26% [54]. Based on this finding, bardoxolone methyl was investigated as a potential treatment for CKD. The earliest study evaluated 20 patients with stage 3b–4 chronic kidney disease (eGFR range 15–45 mL/min/ 1.73 m²) due to diabetes. In this non-placebo-controlled trial, eGFR statistically increased after 4 and 8 weeks compared with baseline (2.8 mL/min/1.73 m² and 7.2 mL/min/1.73 m² at each time point, respectively) [55].

This study was followed by the BEAM study—a double-blinded, placebo-controlled trial designed to assess the efficacy and safety of bardoxolone methyl in patients with Stage 3b to 4 diabetic kidney disease. Of the 227 patients enrolled, 57 received placebo, and the remainder were evenly divided into three different doses of bardoxolone methyl (25 mg, 75 mg, and 150 mg). After one year, the change in eGFR was significantly higher in the bardoxolone methyl groups as compared with placebo (5.8 ± 1.8 mL/min/1.73 m² for 25 mg vs. placebo, 10.5 ± 1.7 mL/min/1.73 m² for 75 mg vs. placebo, and 9.3 ± 1.9 for 150 mg ml/min/1.73 m² vs. placebo) [56].

The BEACON trial was a phase 3 double-blinded, placebo-controlled trial in individuals with stage 4 diabetic kidney disease (eGFR range 15–29 mL/min/1.73 m²) designed to test the hypothesis that bardoxolone methyl would reduce the risk of end-stage kidney disease or death from cardiovascular causes in these patients [57]. The trial was stopped early due to an increased incidence of hospitalization or death from heart failure in the bardoxolone methyl group [57]. After a median of nine months of follow-up, however, the bardoxolone methyl group did show an increase in eGFR of 5.5 mL/min/1.73 m² vs. −0.9 mL/min/1.73 m² for the placebo group [57]. Post hoc analysis identified those at risk for fluid overload [58], and the TSUBAKI study evaluated whether bardoxolone methyl would increase eGFR in patients with diabetic kidney disease in whom these risk factors were absent [59]. In this study, 85 patients with stage 3 or 4 diabetic kidney disease (eGFR range 15–60 mL/min/1.73 m²), and with no identified risk factors for heart failure, were randomized to receive bardoxolone methyl or placebo. After 16 weeks of follow-up, the bardoxolone methyl group showed an increase in inulin-measured GFR of 5.95 mL/min/1.73 m² vs. −0.69 mL/min/1.73 m² for the placebo group.

The role of bardoxolone methyl in nondiabetic kidney disease is currently being evaluated in the CARDINAL study of Alport’s syndrome. Although not published at the time of this review, a press release from Reata pharmaceuticals highlighted the year one results: at 48 weeks of treatment, patients treated with bardoxolone had a statistically significant improvement in mean eGFR of 9.50 mL/min/1.73 m² (p < 0.0001) compared to placebo [60].

Taken together, these data suggest that pharmacologic intervention in the KEAP1/NRF2 pathway can increase GFR. Whether this leads to a reduction of progression to end-stage kidney disease is still not certain, and the negative cardiovascular effects are still a concern. Further, despite the increase in GFR, bardoxolone methyl seems to lead to an increase in urinary protein excretion [57,59], raising the question and concern whether this could be related to the undesirable glomerular hyperfiltration or increased intraglomerular pressure that contributes to kidney disease progression long term [61]. An alternative explanation is that bardoxolone may have a favorable effect on glomerular surface area, and the increase in albuminuria may be tubular rather than glomerular in origin [62]. Animal studies suggest that bardoxolone has a narrow therapeutic window, since its metabolites could also be toxic [63]. Thus, the jury is still out regarding the long-term effect of bardoxolone on kidney function and disease progression.