Molecular basis of the Keap1–Nrf2 system

Takafumi Suzuki, Masayuki Yamamoto n

Department of Medical Biochemistry, Tohoku University Graduate School of Medicine, Sendai 980-8575, Japan

a r t i c l e i n f o

Article history:Received 10 May 2015Received in revised form15 June 2015Accepted 15 June 2015Available online 25 June 2015

Keywords:Nrf2Keap1Stress responseCarcinogenesisChemical inducersFree radicals

a b s t r a c t

Nrf2 (NF-E2-related factor 2) is a master regulator of cellular responses against environmental stresses.Nrf2 induces the expression of detoxification and antioxidant enzymes, and Keap1 (Kelch-like ECH-as-sociated protein 1), an adaptor subunit of Cullin 3-based E3 ubiquitin ligase, regulates Nrf2 activity.Keap1 also acts as a sensor for oxidative and electrophilic stresses. Keap1 retains multiple sensor cysteineresidues that detect various stress stimuli. Increasing attention has been paid to the roles that Nrf2 playsin the protection of our bodies against drug toxicity and stress-induced diseases. On the other hand, Nrf2is found to promote both oncogenesis and cancer cell resistance against chemotherapeutic drugs. Thus,although Nrf2 acts to protect our body from deleterious stresses, cancer cells hijack the Nrf2 activity tosupport their malignant growth. Nrf2 has emerged as a new therapeutic target, and both inducers andinhibitors of Nrf2 are awaited. Studies challenging the molecular basis of the Keap1–Nrf2 system func-tions are now critically important to improve translational studies of the system. Indeed, recent studiesidentified cross talk between Nrf2 and other signaling pathways, which provides new insights into themechanisms by which the Keap1–Nrf2 system serves as a potent regulator of our health and disease.

& 2015 Published by Elsevier Inc.

Contents

1. In vivo validation of Keap1 as a negative regulator of Nrf2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 942. Molecular mechanisms of stress sensing by Keap1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943. Two types of Nrf2 inducers as drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954. Potential target diseases of Nrf2 inducers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955. Roles Nrf2 plays in oncogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 966. Nrf2 inducers and carcinogenesis and metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 977. Cross talk between Nrf2 and other signaling pathways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 978. Concluding remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98

The defense system of our body is equipped with the capacityto upregulate the expression levels of cytoprotective enzymegenes. Nrf2 (nuclear factor erythroid 2-related factor 2) is the mainplayer in the inducible expression of our cellular defense enzymes[1,2]. Nrf2 belongs to the cap-n-collar subfamily of basic region–leucine zipper-type transcription factors [3]. Nrf2 dimerizeswith one of the small Maf proteins and binds to antioxidant/electrophile-response elements located in the regulatory regions

of many defense enzyme genes [1]. Several hundred Nrf2 targetgenes have been identified through gene expression profilinganalysis and chromatin immunoprecipitation analysis, exploitingthe Nrf2 gene-knockout mice as a reference [4]. The Nrf2 targetgenes encode enzymes involved in the synthesis and conjugationof glutathione, antioxidant enzymes, drug-metabolizing enzymes,transporters, and pentose phosphate pathway enzymes [4–6].Thus, Nrf2 activates a wide range of cellular defense processes,thereby enhancing the overall capacity of cells to detoxify andeliminate harmful substances. This review will focus on the mo-lecular basis of the Kelch-like ECH-associated protein 1 (Keap1)–

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/freeradbiomed

Free Radical Biology and Medicine

http://dx.doi.org/10.1016/j.freeradbiomed.2015.06.0060891-5849/& 2015 Published by Elsevier Inc.

n Corresponding author. Fax: þ81 22 717 8090.E-mail address: masiyamamoto@med.tohoku.ac.jp (M. Yamamoto).

Free Radical Biology and Medicine 88 (2015) 93–100

Nrf2 system function, which is now critically important for thetranslation study of this system.

1. In vivo validation of Keap1 as a negative regulator of Nrf2

Keap1 has been identified as a factor interacting with the Neh2(Nrf2-ECH homology domain 2) degron domain of Nrf2 [2]. Keap1is an adaptor subunit of a Cullin 3 (Cul3)-based ubiquitin E3 ligase[7]. Under unstressed conditions, Keap1 binds to Nrf2 in the cy-toplasm and promotes the ubiquitination and proteasomal de-gradation of Nrf2. Upon exposure to chemicals (often electro-philes) or reactive oxygen species (ROS), the ubiquitin E3 ligaseactivity of the Keap1–Cul3 complex declines, and Nrf2 is stabilized.The stabilized Nrf2 accumulates in the nucleus and activates itstarget genes.

The Keap1 function as a negative regulator of Nrf2 has beenvalidated through experiments utilizing mouse genetics. Whilestraightforward Keap1-knockout mice die at the time of weaningowing to hyperkeratosis of the upper digestive tract, which leadsto feeding problems [8,9], analyses exploiting the knockdown andother hypomorphic lines of mice and mouse embryonic fibroblasts(MEFs) revealed that Nrf2 accumulates in the nucleus and cyto-protective genes are constitutively upregulated in various tissuesin the Keap1-deficient mice and in MEFs. The hyperkeratosis andthe massive induction of the expression of cytoprotective genesare reversed in Keap1::Nrf2 compound-knockout mice, indicating

that the phenotype of Keap1-deficient mice is attributable to theconstitutive stabilization of Nrf2. These results support the con-tention that Keap1 is the important negative regulator of Nrf2in vivo.

2. Molecular mechanisms of stress sensing by Keap1

From the beginning of the Keap1–Nrf2 system analyses, thissystem has been linked to a group of biological inducers, collec-tively named as Nrf2 inducers [10,11]. It has been found that themajority of the chemical Nrf2 inducers are electrophilic and reactwith nucleophilic thiols, including cysteine sulfhydryl groups [12].Several mass spectrometry analyses have identified specific pat-terns of Keap1 cysteine modifications by individual Nrf2-inducingchemicals [13,14]. The modification of Keap1 cysteine residuesresults in the inhibition of the ubiquitin E3 ligase activity of theKeap1–Cul3 complex.

The functional significance of these cysteine residues of Keap1has been examined in experiments exploiting site-directed mu-tagenesis [15–17]. The results suggest that the different chemicalsthat trigger the Keap1–Nrf2 system are associated with distinctpatterns of Keap1 cysteine modification. Unique utilization of thesensor cysteine residues leads to the “cysteine code” concept. Thecysteine code appears to support the unique Keap1–Nrf2 systemfunction that responds to a diverse array of chemicals and oxida-tive insults [4].

Fig. 1. The interaction of Keap1 and Nrf2. (A) In response to oxidative and electrophilic stresses, Nrf2 induces the expression of cytoprotective enzyme genes. Underunstressed conditions, Nrf2 is degraded in a Keap1-dependent manner via the ubiquitin (Ub)–proteasome pathway. The Keap1 homodimer binds to single Nrf2 molecules bymeans of two-site binding utilizing DLGex and ETGE motifs. Both motifs individually bind to a pocket in the DC (double glycine repeat and C-terminal) domain of Keap1.Lysine (K) residues that reside between the two motifs are the targets of ubiquitination. Nrf2 inducers inactivate Keap1 via the modification of cysteine residues (Cys), andNrf2 is stabilized, and de novo synthesized Nrf2 translocates into the nucleus. Nrf2 heterodimerizes with small Maf proteins (sMaf) and activates target genes throughbinding to the antioxidant response element (ARE), exerting cytoprotective effects against various noxious insults. (B) Phylogenetic alignment of N-terminal Neh2 domain ofNrf2. Conserved amino acid residues are shaded in black. DLGex and ETGE motifs are indicated with pink and red bars, respectively. Three helices in DLGex motif andantiparallel β-sheet in the ETGE motif are indicated with black bars and arrows, respectively.

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–10094

It has been explored how environmental stresses are trans-mitted to the gene expression regulation. We have identified Keap1as a sensor of electrophiles and ROS, so that the remaining questionis how signals detected by Keap1 are transmitted for Nrf2 activationand induction of cytoprotective gene expression. One unique me-chanism proposed for the Keap1-mediated activation of Nrf2 is the“hinge and latch model” [18]. This model has been proposed basedon the finding that Keap1 homodimer binds to a single Nrf2 mo-lecule through two distinct binding sites (i.e., DLG and ETGE motifs)within the Neh2 domain of Nrf2. As shown in Fig. 1A, both motifsindividually bind to a pocket within the DC (double glycine repeatand C-terminal) domain of Keap1 [19]. Binding of the high-affinityETGE motif and low-affinity DLG motif provides a hinge and latch,which facilitate optimal positioning of the lysine residues betweenthe two motifs for ubiquitin conjugation.

Fukutomi and colleagues recently identified that the DLG motifneeds to be much more extended (DLGex motif (Met17–Gln51))than those assumed by the classical DLG motif and DIDLID ele-ment [20]. The DLGex possesses triple helices, Helix-1 (Leu19–Arg25), Helix-2 (Ile28–Leu30), and Helix-3 (Arg34–Phe37)(Fig. 1B). This DLGex structure shows striking differences from theβ-hairpin structure of ETGE [21]. The binding modes of DLGex andETGE to Keap1 are quite distinct from each other. The Keap1–DLGex binding is thermodynamically characterized as both en-thalpy and entropy driven and kinetically as fast-on and fast-off. Incontrast, the ETGE–Keap1 binding is characterized as purely en-thalpy driven and contains a two-state reaction that leads to astable conformation [20]. These observations support the con-tention that the DLGex motif serves as a converter transmittingenvironmental stress to Nrf2 induction as the latch site.

On the other hand, the open and closed states of the Keap1–Nrf2 complex were studied by fluorescence resonance energytransfer experiments [22]. This study showed that chemical Nrf2inducers rather strengthen the Keap1–DLGex binding and pro-mote the closed conformation of the Keap1–Nrf2 complex. In thismodel the latch site becomes too tight and, as a result, Nrf2 isstabilized. Similar observations have been reported through the

cancer somatic mutation (superbinder) analyses [23]. Althoughthese models do not show immediate consistency (nor dis-crepancy) with the hinge and latch model, the importance of thetwo-site binding and the DLG motif participation in the regulationare in common. Further investigations are necessary to verify howthese models explain the accumulation of Nrf2.

In this regard, another important observation is that cysteine-151, one of the critical cysteine residues, is localized far away fromthe latch site [24]. Available evidence suggests that modification ofthis residue may interfere with the interaction of Keap1 with Cul3[25–27].

3. Two types of Nrf2 inducers as drugs

There is substantial interest in developing Nrf2 inducers fortherapeutic use. One of the most promising chemical Nrf2 inducersis CDDO (2-cyano-3,12-dioxooleana-1,9(11)-diene-28-oic acid) tri-terpenoids derived from oleanolic acid, which has antioxidant andanti-cancer properties [28]. The methyl ester derivative (CDDO-Me)of the triterpenoid compound is a potent inducer of Nrf2. CDDO-Merobustly stimulates the expression of Nrf2-dependent cytoprotec-tive enzyme genes at low nanomolar concentrations. CDDO-Me hasbeen studied in clinical trials, under the generic name bardoxolonemethyl, to assess its potential for the treatment of a variety of dis-orders. In particular, a phase II clinical trial for chronic kidney dis-ease (BEAM study) showed promising results [29]. Unfortunately, aphase III trial of bardoxolone (BEACON study) was terminated inOctober of 2012 owing to a higher rate of adverse events [30].However, CDDO-Me is now under clinical investigation for thetreatment of pulmonary hypertension in the United States (https://clinicaltrials.gov/ct2/show/NCT02036970) and for chronic kidneydisease associated with type 2 diabetes in Japan (http://www.kyowa-kirin.com/news_releases/2013/e20131111_01.html). Another at-tractive Nrf2 inducer is dimethyl fumarate (Tecfidera), which hasbeen approved for the treatment of relapsing multiple sclerosis bythe U.S. Food and Drug Administration (http://www.fda.gov/Drugs/DrugSafety/ucm424625.htm). Whereas the exact mechanisms bywhich dimethyl fumarate exerts its clinical efficacy remain to beclarified, the effects are believed, at least in part, to be mediatedthrough Nrf2 activation.

As shown in Fig. 2, most of the chemical Nrf2 inducers. in-cluding the above two. are known to interact with cysteine re-sidues of Keap1 and inactivate the Keap1 E3 ligase activity. Be-cause the classic Nrf2 inducers interact with the cysteine residuesof Keap1 by utilizing their electrophilic nature, these inducers ofNrf2 inherently retain the ability to react with intracellular glu-tathione or thiol of proteins. Therefore, overdose of the com-pounds potentially gives rise to toxic effects. In this regard, che-micals directly interrupting the interaction of Keap1 and Nrf2 havebeen emerging as attractive new inducers of the Nrf2 activity(Fig. 2). These nonelectrophilic inducers exert their effect withoutcausing electrophilic damage to cells. Two high-throughputscreenings searching for small-molecular inhibitors of the Keap1–Nrf2 interaction have been reported. One is a fluorescence polar-ization-based screening that has identified (SRS)-5 [31], and theother is the homogeneous confocal fluorescence anisotropy assay-based screening that gave rise to Cpd16 [32]. Whereas thesenonelectrophilic compounds are expected to become safe chemi-cal Nrf2 inducers, further improvements are apparently needed toachieve low nanomolar efficacy.

4. Potential target diseases of Nrf2 inducers

Nrf2-knockout (Nrf2" /") mice are inherently more susceptibleto oxidative stress-based diseases and drug-induced toxicity. Many

Fig. 2. Two types of Nrf2 chemical inducers. Most of the chemical Nrf2 inducersinteract with cysteine residues (Cys) of Keap1 by utilizing the electrophilic natureof the molecules and inactivating the Keap1 E3 ligase activity. This process is re-fereed to as the cysteine code. The other type of Nrf2 inducer is nonelectrophilicinducers, which interrupt the interaction between Keap1 and Nrf2 directly andinactivate the Keap1 E3 ligase activity. Nonelectrophilic inducers such as Cpd16 and(SRS)-5 probably disrupt only the binding of the DLG motif to Keap1 because thebinding affinities of these inducers are weaker than the binding affinity of the ETGEmotif.

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–100 95

lines of experiments demonstrate that Nrf2 activation by phar-macological (i.e., pretreatment with chemical Nrf2 inducers) orgenetic (i.e., the disruption of the Keap1 gene) approaches sig-nificantly reduces cellular damage caused by oxidative stress andprevents development and exacerbation of stress-induced dis-eases. For example, as shown in Figs. 3, 4-nitroquinoline-1-oxide(4NQO)-induced carcinogenesis of the upper aerodigestive tract isnicely prevented by genetic Nrf2 induction [33]. This cancer-pre-ventive effect seem to be due to the high-level expression of Nrf2and subsequent expression of detoxifying enzymes specific for4NQO. The detoxification of 4NQO is mostly mediated by glu-tathione conjugation [34,35]. Of the Nrf2 target genes, glutamate–cysteine ligase catalytic subunit increases glutathione synthesisand glutathione S-transferase P catalyzes the formation of 4NQO–glutathione [35]. In addition, thickened stratified squamous epi-thelium appears to protect the tissues against exposure to carci-nogens although it has not been examined whether tissue levels of4NQO are affected by Nrf2 activation (Fig. 3).

Another interesting topic that has emerged as a target disease ofNrf2 inducers is diabetes mellitus. Accumulating lines of evidencedemonstrate that Nrf2 activation brings about the suppression ofdiabetes mellitus. This antidiabetic effect of Nrf2 inducers can beexplained by two distinct mechanisms. First, the Nrf2 inducersprotect pancreatic β-cells against oxidative damage [36]. Second, theactivation of Nrf2 suppresses the onset and/or progression of insulinresistance in skeletal muscle and liver [37]. These two functions ofNrf2 together lead to the protection against diabetes [38]. In contrastto the above observations of Nrf2 gain-of-function studies, severalreports have shown a decrease in insulin resistance in Nrf2-knock-out mice [39–41]. Although the molecular basis for these bidirec-tional effects of Nrf2 on insulin resistance (i.e., both gain and loss ofNrf2 function reduce insulin resistance) awaits further studies, wesurmise that the increase in insulin sensitivity brought by gain ofNrf2 function may be explained by the suppression of gluconeo-genesis [37], Fgf21 elevation [42], and AMPK activation [43,44], andthe increase in insulin sensitivity by loss of Nrf2 function may beexplained by enhancement of insulin signaling owing to increased

H2O2 levels [41].

5. Roles Nrf2 plays in oncogenesis

It has been shown that Nrf2" /" mice are susceptible to variouscarcinogens [33,45–50]. Showing very good agreement with theobservations utilizing Nrf2" /" mice, a regulatory single-nucleotidepolymorphism (rSNP) in the human Nrf2 gene upstream promoterregion presents an intriguing correlation between loss of Nrf2expression and increase in carcinogenesis. At 617 bases upstreamfrom the transcription start site, the major allele is C/C and theminor allele A/A (rs6721961) [51,52]. Minor A/A homozygotesexhibit significantly diminished Nrf2 expression in peripheralblood lymphocytes in vivo [53] and in reporter cotransfectiontransactivation assays [52]. When a cohort of lung cancer cases

Fig. 3. Roles the Keap1–Nrf2 system plays in upper aerodigestive tract carcinogenesis by 4NQO. In the absence of Nrf2 (Nrf2 KO), 4NQO dramatically induced carcinogenesis,which is probably due to reduced basal expression of detoxifying enzymes and a relatively thin stratified epithelium. When Keap1 is knocked down (Keap1 KD), Nrf2 isconstitutively activated and 4NQO-induced carcinogenesis is significantly inhibited. In addition to the upregulation of detoxifying enzymes, increase in growth-relatedmetabolic enzyme genes leads to increased progenitor cell growth and thickened stratified squamous epithelium with the keratin layer, which may serve as a mechanicaldefense.

Fig. 4. Association of Nrf2 rSNP and increased risk and reduced malignancy of lungcancers. A/A homozygotes for a Nrf2 rSNP at "617 exhibit significantly decreasedexpression of Nrf2 and downstream cytoprotective enzyme genes, resulting inenhanced initiation of carcinogenesis, but reduced malignancy of cancer cells.

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–10096

was investigated, the A/A homozygotes showed a high incidence ofnon-small-cell lung carcinomas, especially in those who had eversmoked [53].

In contrast, analyses of a urethane-induced multistep lungcarcinogenesis model exploiting Nrf2"/" mice as references giverise to distinct results [54,55]. Satoh and colleagues have provedthat Nrf2 has dual roles in urethane carcinogenesis. One is toprevent tumor initiation, and the other is to promote malignantprogression of the tumors [55]. Consistent with the mouse ob-servations, lung cancer patients carrying the minor A/A allele at"617 exhibit remarkable survival after surgical operation [56].These observations demonstrate that Nrf2 plays distinct roles inthe initiation and promotion of cancer (Fig. 4).

Elevations in Nrf2 levels have been identified in various typesof human cancers. Somatic mutations, including missense muta-tions, insertions, and deletions, have been identified in both Nrf2and Keap1 genes of various cancers [21,57–60]. Importantly, thesemutations are quite often associated with the stabilization of Nrf2.Mutations identified within the Nrf2 gene are clustered within orin proximity to the DLG and ETGE motifs in the Neh2 domain,which results in the disruption of Nrf2–Keap1 interactions andsubsequently the stabilization of Nrf2. This natural evidencestrongly supports the contention that the integrity of the two-sitebinding motifs in Nrf2 is essential for Nrf2 degradation and furthersupport the hinge and latch model.

In addition to the somatic mutations in the Keap1 and Nrf2genes, several other mechanisms have been found to provoke Nrf2stabilization. For instance, promoter methylation of the Keap1gene suppresses expression of the gene and results in Nrf2 acti-vation [61]. Fumarate hydratase inactivation in renal carcinomasresults in Nrf2 activation via oncometabolite modification (succi-nation) of the Keap1 cysteine residues by accumulated fumarate,although the contribution of Nrf2 activation upon cancer malig-nancy remains unknown [62,63]. Transcriptional activation of Nrf2by the Ras oncogenic pathway also contributes to the enhance-ment of Nrf2 activity in cancer [64]. Various factors can disrupt thenormal regulation of the Keap1–Nrf2 pathway, and these are in-deed associated with carcinogenesis and/or malignanttransformation.

Considering the cytotoxic properties of anti-cancer drugs andNrf2’s ability to activate the expression of detoxification enzymes,antioxidant enzymes/proteins, and xenobiotic transporters, itseems reasonable to assume that Nrf2 activation enhances cancercells’ resistance to various chemoradiotherapies [65]. Recently,another line of evidence demonstrating the advantage of con-stitutive Nrf2 activation for cancer cell proliferation was identified.In rapidly growing cancer cells, Nrf2 acts to redirect glucose andglutamine into the anabolic pathway and support cell proliferation[66]. Nrf2 also activates the pentose phosphate pathway, indicat-ing that Nrf2 activation heavily commits to the oncogenictransformation.

6. Nrf2 inducers and carcinogenesis and metastasis

Upon clinical applications of chemical Nrf2 inducers, adverseeffects need to be considered. There is a growing concern that Nrf2inducers may confer resistance to cancer cells that incidentallyreside in patients. In this regard, we surmise that accumulatinglines of evidence utilizing Keap1 and Nrf2 gene-modified mice willgive rise to impartial views on this issue.

Intriguing observations are that treatment of mice with potentNrf2 inducers rather decelerates cancer development in certainmouse models of prostate, breast, and pancreatic cancers [67–69].CDDO-Me shows promising preventive and therapeutic activitiesagainst certain types of cancers, not only in animal studies but also

in clinical trials [70]. The direct anti-tumor activity of CDDO-Mecan be explained by inhibition of PI3K (phosphoinositide 3-kinase)and nuclear factor κB signaling rather than Nrf2 activation [71].CDDO-Me is also found to abrogate the immune-suppressive ef-fects of myeloid-derived suppressor cells (MDSCs) and improveimmune responses in cancer patients [72].

It has been shown that Nrf2 exerts a protective function againstcancer cell metastasis [73]. Upon intravenous injection of Lewislung carcinoma (3LL) cells, Nrf2" /" mice show a significantly in-creased number of metastatic nodules in the lung, which is asso-ciated with aberrant ROS accumulation in MDSCs. In contrast, Nrf2activation by CDDO-Im reduces the ROS level in MDSCs and ef-fectively prevents 3LL cancer cell metastasis. This prevention ofmetastasis is cancelled by myeloid-lineage-specific knockout ofNrf2 [74], indicating that Nrf2 activation in MDSCs prevents cancercell metastasis.

These wide-ranging observations suggest that the originallyproposed therapeutic benefits brought about by activating Nrf2probably exceed the exacerbation risk of Nrf2 activation in phar-macological treatments of various diseases (Fig. 5). Furthermore,Nrf2 inducers may be applicable for cancer chemotherapy. Wesurmise that Nrf2 inducers will reinforce cancer chemotherapy ifthe improvement in the microenvironment slows down thegrowth of cancer cells.

7. Cross talk between Nrf2 and other signaling pathways

There are a number of reports describing cross talk betweenNrf2 and other signaling pathways. The Keap1-independent de-gradation pathway mediated by the Cul1–β-TrCP (β-transducinrepeat-containing protein)-based ubiquitin E3 ligase has been re-ported [75]. This Cul1–β-TrCP-mediated degradation of Nrf2 iscontrolled by phosphorylation of the Neh6 domain of Nrf2, whichcan be regulated by GSK-3 [76]. In addition, deletion of PTEN(phosphatase and tensin homolog deleted from chromosome 10)enhances Nrf2 accumulation in Keap1-deficient liver [77], sug-gesting that sustained activation of the PI3K–Akt pathway andinactivation of GSK-3 by the loss of PTEN inhibits Nrf2 degradationvia the Cul1–β-TrCP pathway. Interestingly, in long-lived nakedmole rats, high basal levels of Nrf2 activity and low expression ofKeap1 and β-TrCP are observed [78]. Similarly, loss of PTEN activitycorrelates well with high Nrf2 protein levels in human en-dometrioid tumors [79]. Whereas these observations suggest thatthe PTEN–GSK-3–β-TrCP pathway is an important regulatory axisfor Nrf2 activity, mouse genetic experiments utilizing Nrf2, Keap1,and PTEN knockout lines of mice in combination provide evidence

Fig. 5. Therapeutic benefits by activating Nrf2 exceed exacerbation risk by acti-vating Nrf2 in cancer cells. Nrf2 inducers potentiate Nrf2 activity in MDSCs (mye-loid-derived suppressor cells) and attenuate ROS accumulation by activating theantioxidant capacity, resulting in the repression of cancer cell metastasis.

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–100 97

that the PTEN–GSK-3–β-TrCP pathway exerts effects supplemen-tary to the Keap1 pathway [77].

Another important cross talk recently identified is between theKeap1–Nrf2 system and the cellular autophagy pathway. Autop-hagy-adaptor protein p62 is found to compete with Keap1 forbinding with Nrf2 by using the STGE motif, and through thiscompetition p62 promotes the stabilization of Nrf2 and upregu-lation of Nrf2 activity [80]. A key finding here is that phosphor-ylation of the STGE motif in p62 results in strong binding to Keap1.The p62 harboring phosphorylated STGE shows markedly in-creased binding affinity to Keap1, almost similar to ETGE, andblocks binding of the DLG motif [81]. Persistent activation of Nrf2through accumulation of phosphorylated p62 contributes to thegrowth of human hepatocellular carcinomas [82].

An intriguing finding in this regard is that, whereas Nrf2 ispredominantly degraded through the proteasome pathway, Keap1is degraded through the autophagy pathway [83]. The autophagypathway maintains the integrity and homeostasis of the Keap1–Nrf2 system by governing Keap1 turnover. Thus, the Keap1–Nrf2system is regulated coordinately through the two important cel-lular proteolysis pathways.

8. Concluding remarks

As summarized above, there have been significant advances inour understanding of the function and regulation of the Keap1–Nrf2system. One of the important advances is the clarification of dif-ferences in the binding modes of the DLG and ETGE motifs to Keap1,although precise elucidation of the stress response mechanismsneeds further study. Based on the current knowledge, chemical Nrf2inducers that interrupt directly the Keap1–Nrf2 interaction havebeen emerging as an attractive strategy. Another important advanceis the identification of the roles that Nrf2 signaling plays in onco-genesis. Nrf2 activity is deeply and differentially involved in theinitiation, promotion, and metastasis of cancer cells.

Acknowledgments

This work was supported in part by MEXT/JSPS KAKENHI(24249015 to M.Y. and 26460354, 25112502, and 2611010 to T.S.),CREST, JST (to M.Y.), P-DIRECT, MEXT) (to T.S. and M.Y.), the NaitoFoundation (to M.Y.), and the Takeda Science Foundation (to M.Y.).

References

[1] K. Itoh, T. Chiba, S. Takahashi, T. Ishii, K. Igarashi, Y. Katoh, T. Oyake, N. Hayashi,K. Satoh, I. Hatayama, M. Yamamoto, Y. Nabeshima, An Nrf2/small Maf het-erodimer mediates the induction of phase II detoxifying enzyme genesthrough antioxidant response elements, Biochem. Biophys. Res. Commun. 236(1997) 313–322.

[2] K. Itoh, N. Wakabayashi, Y. Katoh, T. Ishii, K. Igarashi, J.D. Engel, M. Yamamoto,Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2through binding to the amino-terminal Neh2 domain, Genes Dev. 13 (1999)76–86.

[3] H. Motohashi, M. Yamamoto, Nrf2-Keap1 defines a physiologically importantstress response mechanism, Trends Mol. Med. 10 (2004) 549–557.

[4] T. Suzuki, H. Motohashi, M. Yamamoto, Toward clinical application of theKeap1-Nrf2 pathway, Trends Pharmacol. Sci. 34 (2013) 340–346.

[5] K. Itoh, T. Ishii, N. Wakabayashi, M. Yamamoto, Regulatory mechanisms ofcellular response to oxidative stress, Free Radic. Res. 31 (1999) 319–324.

[6] J.D. Hayes, A.T. Dinkova-Kostova, The Nrf2 regulatory network provides aninterface between redox and intermediary metabolism, Trends Biochem. Sci.39 (2014) 199–218.

[7] A. Kobayashi, M.I. Kang, H. Okawa, M. Ohtsuji, Y. Zenke, T. Chiba, K. Igarashi,M. Yamamoto, Oxidative stress sensor Keap1 functions as an adaptor for Cul3-based E3 ligase to regulate proteasomal degradation of Nrf2, Mol. Cell. Biol. 24(2004) 7130–7139.

[8] N. Wakabayashi, K. Itoh, J. Wakabayashi, H. Motohashi, S. Noda, S. Takahashi,

S. Imakado, T. Kotsuji, F. Otsuka, D.R. Roop, T. Harada, J.D. Engel, M. Yamamoto,Keap1-null mutation leads to postnatal lethality due to constitutive Nrf2 ac-tivation, Nat. Genet. 35 (2003) 238–245.

[9] K. Taguchi, J.M. Maher, T. Suzuki, Y. Kawatani, H. Motohashi, M. Yamamoto,Genetic analysis of cytoprotective functions supported by graded expression ofKeap1, Mol. Cell. Biol. 30 (2010) 3016–3026.

[10] T.W. Kensler, N. Wakabayashi, Nrf2: friend or foe for chemoprevention? Car-cinogenesis 31 (2010) 90–99.

[11] A.T. Dinkova-Kostova, R.V. Kostov, Glucosinolates and isothiocyanates inhealth and disease, Trends Mol. Med. 18 (2012) 337–347.

[12] A.T. Dinkova-Kostova, P. Talalay, Direct and indirect antioxidant properties ofinducers of cytoprotective proteins, Mol. Nutr. Food Res. 52 (Suppl. 1) (2008)S128–138.

[13] A.T. Dinkova-Kostova, W.D. Holtzclaw, R.N. Cole, K. Itoh, N. Wakabayashi,Y. Katoh, M. Yamamoto, P. Talalay, Direct evidence that sulfhydryl groups ofKeap1 are the sensors regulating induction of phase 2 enzymes that protectagainst carcinogens and oxidants, Proc. Natl. Acad. Sci. USA 99 (2002)11908–11913.

[14] M. Kobayashi, L. Li, N. Iwamoto, Y. Nakajima-Takagi, H. Kaneko, Y. Nakayama,M. Eguchi, Y. Wada, Y. Kumagai, M. Yamamoto, The antioxidant defense sys-tem Keap1-Nrf2 comprises a multiple sensing mechanism for responding to awide range of chemical compounds, Mol. Cell. Biol. 29 (2009) 493–502.

[15] D.D. Zhang, M. Hannink, Distinct cysteine residues in Keap1 are required forKeap1-dependent ubiquitination of Nrf2 and for stabilization of Nrf2 by che-mopreventive agents and oxidative stress, Mol. Cell. Biol. 23 (2003)8137–8151.

[16] T. Yamamoto, T. Suzuki, A. Kobayashi, J. Wakabayashi, J. Maher, H. Motohashi,M. Yamamoto, Physiological significance of reactive cysteine residues of Keap1in determining Nrf2 activity, Mol. Cell. Biol. 28 (2008) 2758–2770.

[17] K. Takaya, T. Suzuki, H. Motohashi, K. Onodera, S. Satomi, T.W. Kensler,M. Yamamoto, Validation of the multiple sensor mechanism of the Keap1-Nrf2system, Free Radic. Biol. Med. 53 (2012) 817–827.

[18] K.I. Tong, A. Kobayashi, F. Katsuoka, M. Yamamoto, Two-site substrate re-cognition model for the Keap1-Nrf2 system: a hinge and latch mechanism,Biol. Chem. 387 (2006) 1311–1320.

[19] K.I. Tong, B. Padmanabhan, A. Kobayashi, C. Shang, Y. Hirotsu, S. Yokoyama,M. Yamamoto, Different electrostatic potentials define ETGE and DLG motifs ashinge and latch in oxidative stress response, Mol. Cell. Biol. 27 (2007)7511–7521.

[20] T. Fukutomi, K. Takagi, T. Mizushima, N. Ohuchi, M. Yamamoto, Kinetic, ther-modynamic, and structural characterizations of the association between Nrf2-DLGex degron and Keap1, Mol. Cell. Biol. 34 (2014) 832–846.

[21] B. Padmanabhan, K.I. Tong, T. Ohta, Y. Nakamura, M. Scharlock, M. Ohtsuji, M.I. Kang, A. Kobayashi, S. Yokoyama, M. Yamamoto, Structural basis for defectsof Keap1 activity provoked by its point mutations in lung cancer, Mol. Cell 21(2006) 689–700.

[22] L. Baird, D. Llères, S. Swift, A.T. Dinkova-Kostova, Regulatory flexibility in theNrf2-mediated stress response is conferred by conformational cycling of theKeap1-Nrf2 protein complex, Proc. Natl. Acad. Sci. USA 110 (2013)15259–15264.

[23] B.E. Hast, E.W. Cloer, D. Goldfarb, H. Li, P.F. Siesser, F. Yan, V. Walter, N. Zheng,D.N. Hayes, M.B. Major, Cancer-derived mutations in Keap1 impair Nrf2 de-gradation but not ubiquitination, Cancer Res. 74 (2014) 808–817.

[24] T. Ogura, K.I. Tong, K. Mio, Y. Maruyama, H. Kurokawa, C. Sato, M. Yamamoto,Keap1 is a forked-stem dimer structure with two large spheres enclosing theintervening, double glycine repeat, and C-terminal domains, Proc. Natl. Acad.Sci. USA 107 (2010) 2842–2847.

[25] D.D. Zhang, S.C. Lo, J.V. Cross, D.J. Templeton, M. Hannink, Keap1 is a redox-regulated substrate adaptor protein for a Cul3-dependent ubiquitin ligasecomplex, Mol. Cell. Biol. 24 (2004) 10941–10953.

[26] L. Gao, J. Wang, K.R. Sekhar, H. Yin, N.F. Yared, S.N. Schneider, S. Sasi, T.P. Dalton, M.E. Anderson, J.Y. Chan, J.D. Morrow, M.L. Freeman, Novel n-3 fattyacid oxidation products activate Nrf2 by destabilizing the association betweenKeap1 and Cullin3, J. Biol. Chem. 282 (2007) 2529–2537.

[27] G. Rachakonda, Y. Xiong, K.R. Sekhar, S.L. Stamer, D.C. Liebler, M.L. Freeman,Covalent modification at Cys151 dissociates the electrophile sensor Keap1from the ubiquitin ligase CUL3, Chem. Res. Toxicol. 21 (2008) 705–710.

[28] K.T. Liby, M.B. Sporn, Synthetic oleanane triterpenoids: multifunctional drugswith a broad range of applications for prevention and treatment of chronicdisease, Pharmacol. Rev. 64 (2012) 972–1003.

[29] P.E. Pergola, P. Raskin, R.D. Toto, C.J. Meyer, J.W. Huff, E.B. Grossman, M. Krauth,S. Ruiz, P. Audhya, H. Christ-Schmidt, J. Wittes, D.G. Warnock, Investigators, B.S. Bardoxolone methyl and kidney function in CKD with type 2 diabetes, N.Engl. J. Med. 365 (2011) 327–336.

[30] D. de Zeeuw, T. Akizawa, P. Audhya, G.L. Bakris, M. Chin, H. Christ-Schmidt,A. Goldsberry, M. Houser, M. Krauth, Heerspink Lambers, H.J. McMurray, J. J.,C.J. Meyer, H.H. Parving, G. Remuzzi, R.D. Toto, N.D. Vaziri, C. Wanner, J. Wittes,D. Wrolstad, G.M. Chertow, Investigators, B. T. Bardoxolone methyl in type2 diabetes and stage 4 chronic kidney disease, N. Engl. J. Med. 369 (2013)2492–2503.

[31] L. Hu, S. Magesh, L. Chen, L. Wang, T.A. Lewis, Y. Chen, C. Khodier, D. Inoyama,L.J. Beamer, T.J. Emge, J. Shen, J.E. Kerrigan, A.N. Kong, S. Dandapani, M. Palmer,S.L. Schreiber, B. Munoz, Discovery of a small-molecule inhibitor and cellularprobe of Keap1-Nrf2 protein-protein interaction, Bioorg. Med. Chem. Lett. 23(2013) 3039–3043.

[32] D. Marcotte, W. Zeng, J.C. Hus, A. McKenzie, C. Hession, P. Jin, C. Bergeron,

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–10098

A. Lugovskoy, I. Enyedy, H. Cuervo, D. Wang, C. Atmanene, D. Roecklin,M. Vecchi, V. Vivat, J. Kraemer, D. Winkler, V. Hong, J. Chao, M. Lukashev,L. Silvian, Small molecules inhibit the interaction of Nrf2 and the Keap1 Kelchdomain through a non-covalent mechanism, Bioorg. Med. Chem. 21 (2013)4011–4019.

[33] A. Ohkoshi, T. Suzuki, M. Ono, T. Kobayashi, M. Yamamoto, Roles of Keap1-Nrf2system in upper aerodigestive tract carcinogenesis, Cancer Prev. Res. (Phila-delphia) 6 (2013) 149–159.

[34] D. Kanojia, M.M. Vaidya, 4-Nitroquinoline-1-oxide induced experimental oralcarcinogenesis, Oral Oncol. 42 (2006) 655–667.

[35] C.S. Morrow, P.K. Smitherman, S.K. Diah, E. Schneider, A.J. Townsend, Co-ordinated action of glutathione S-transferases (GSTs) and multidrug resistanceprotein 1 (MRP1) in antineoplastic drug detoxification: mechanism of GST A1-1- and MRP1-associated resistance to chlorambucil in MCF7 breast carcinomacells, J. Biol. Chem. 273 (1998) 20114–20120.

[36] Y. Yagishita, T. Fukutomi, A. Sugawara, H. Kawamura, T. Takahashi, J. Pi,A. Uruno, M. Yamamoto, . Nrf2 protects pancreatic β-cells from oxidative andnitrosative stress in diabetic model mice, Diabetes 63 (2014) 605–618.

[37] A. Uruno, Y. Furusawa, Y. Yagishita, T. Fukutomi, H. Muramatsu, T. Negishi,A. Sugawara, T.W. Kensler, M. Yamamoto, The Keap1-Nrf2 system preventsonset of diabetes mellitus, Mol. Cell. Biol. 33 (2013) 2996–3010.

[38] A. Uruno, Y. Yagishita, M. Yamamoto, The Keap1-Nrf2 system and diabetesmellitus, Arch. Biochem. Biophys. 566 (2015) 76–84.

[39] A.K. Meher, P.R. Sharma, V.A. Lira, M. Yamamoto, T.W. Kensler, Z. Yan,N. Leitinger, Nrf2 deficiency in myeloid cells is not sufficient to protect micefrom high-fat diet-induced adipose tissue inflammation and insulin re-sistance, Free Radic. Biol. Med. 52 (2012) 1708–1715.

[40] Y.K. Zhang, K.C. Wu, J. Liu, C.D. Klaassen, Nrf2 deficiency improves glucosetolerance in mice fed a high-fat diet, Toxicol. Appl. Pharmacol. 264 (2012)305–314.

[41] P.J. Meakin, S. Chowdhry, R.S. Sharma, F.B. Ashford, S.V. Walsh, R.J. McCrimmon, A.T. Dinkova-Kostova, J.F. Dillon, J.D. Hayes, M.L. Ashford,Susceptibility of Nrf2-null mice to steatohepatitis and cirrhosis upon con-sumption of a high-fat diet is associated with oxidative stress, perturbation ofthe unfolded protein response, and disturbance in the expression of metabolicenzymes but not with insulin resistance, Mol. Cell. Biol. 34 (2014) 3305–3320.

[42] Y. Furusawa, A. Uruno, Y. Yagishita, C. Higashi, M. Yamamoto, Nrf2 inducesfibroblast growth factor 21 in diabetic mice, Genes Cells 19 (2014) 864–878.

[43] P.K. Saha, V.T. Reddy, M. Konopleva, M. Andreeff, L. Chan, The triterpenoid2-cyano-3,12-dioxooleana-1,9-dien-28-oic-acid methyl ester has potent anti-diabetic effects in diet-induced diabetic mice and Lepr(db/db) mice, J. Biol.Chem. 285 (2010) 40581–40592.

[44] J. Xu, A.C. Donepudi, J.E. Moscovitz, A.L. Slitt, Keap1-knockdown decreasesfasting-induced fatty liver via altered lipid metabolism and decreased fattyacid mobilization from adipose tissue, PLoS One 8 (2013) e79841.

[45] M. Ramos-Gomez, M.K. Kwak, P.M. Dolan, K. Itoh, M. Yamamoto, P. Talalay, T.W. Kensler, Sensitivity to carcinogenesis is increased and chemoprotectiveefficacy of enzyme inducers is lost in nrf2 transcription factor-deficient mice,Proc. Natl. Acad. Sci. USA 98 (2001) 3410–3415.

[46] K. Iida, K. Itoh, Y. Kumagai, R. Oyasu, K. Hattori, K. Kawai, T. Shimazui, H. Akaza,M. Yamamoto, Nrf2 is essential for the chemopreventive efficacy of oltiprazagainst urinary bladder carcinogenesis, Cancer Res. 64 (2004) 6424–6431.

[47] C. Xu, M.T. Huang, G. Shen, X. Yuan, W. Lin, T.O. Khor, A.H. Conney, A.N. Kong,Inhibition of 7,12-dimethylbenz(a)anthracene-induced skin tumorigenesis inC57BL/6 mice by sulforaphane is mediated by nuclear factor E2-related factor2, Cancer Res. 66 (2006) 8293–8296.

[48] Y. Kitamura, T. Umemura, K. Kanki, Y. Kodama, S. Kitamoto, K. Saito, K. Itoh,M. Yamamoto, T. Masegi, A. Nishikawa, M. Hirose, Increased susceptibility tohepatocarcinogenicity of Nrf2-deficient mice exposed to 2-amino-3-methyli-midazo[4,5-f]quinoline, Cancer Sci. 98 (2007) 19–24.

[49] T.O. Khor, M.T. Huang, A. Prawan, Y. Liu, X. Hao, S. Yu, W.K. Cheung, J.Y. Chan,B.S. Reddy, C.S. Yang, A.N. Kong, Increased susceptibility of Nrf2 knockout miceto colitis-associated colorectal cancer, Cancer Prev. Res. (Philadelphia) 1 (2008)187–191.

[50] L. Becks, M. Prince, H. Burson, C. Christophe, M. Broadway, K. Itoh,M. Yamamoto, M. Mathis, E. Orchard, R. Shi, J. McLarty, K. Pruitt, S. Zhang, H.E. Kleiner-Hancock, Aggressive mammary carcinoma progression in Nrf2knockout mice treated with 7,12-dimethylbenz[a]anthracene, BMC Cancer 10(2010) 540.

[51] T. Yamamoto, K. Yoh, A. Kobayashi, Y. Ishii, S. Kure, A. Koyama, T. Sakamoto,K. Sekizawa, H. Motohashi, M. Yamamoto, Identification of polymorphisms inthe promoter region of the human Nrf2 gene, Biochem. Biophys. Res. Com-mun. 321 (2004) 72–79.

[52] J.M. Marzec, J.D. Christie, S.P. Reddy, A.E. Jedlicka, H. Vuong, P.N. Lanken,R. Aplenc, T. Yamamoto, M. Yamamoto, H.Y. Cho, S.R. Kleeberger, Functionalpolymorphisms in the transcription factor Nrf2 in humans increase the risk ofacute lung injury, FASEB J 21 (2007) 2237–2246.

[53] T. Suzuki, T. Shibata, K. Takaya, K. Shiraishi, T. Kohno, H. Kunitoh, K. Tsuta,K. Furuta, K. Goto, F. Hosoda, H. Sakamoto, H. Motohashi, M. Yamamoto,Regulatory nexus of synthesis and degradation deciphers cellular Nrf2 ex-pression levels, Mol. Cell. Biol. 33 (2013) 2402–2412.

[54] A.K. Bauer, H.Y. Cho, L. Miller-Degraff, C. Walker, K. Helms, J. Fostel,M. Yamamoto, S.R. Kleeberger, Targeted deletion of Nrf2 reduces urethane-induced lung tumor development in mice, PLoS One 6 (2011) e26590.

[55] H. Satoh, T. Moriguchi, J. Takai, M. Ebina, M. Yamamoto, Nrf2 prevents initia-tion but accelerates progression through the Kras signaling pathway during

lung carcinogenesis, Cancer Res. 73 (2013) 4158–4168.[56] Y. Okano, U. Nezu, Y. Enokida, M.T. Lee, H. Kinoshita, A. Lezhava,

Y. Hayashizaki, S. Morita, M. Taguri, Y. Ichikawa, T. Kaneko, Y. Natsumeda,T. Yokose, H. Nakayama, Y. Miyagi, T. Ishikawa, SNP ("617C4A) in ARE-likeloci of the Nrf2 gene: a new biomarker for prognosis of lung adenocarcinomain Japanese non-smoking women, PLoS One 8 (2013) e73794.

[57] T. Ohta, K. Iijima, M. Miyamoto, I. Nakahara, H. Tanaka, M. Ohtsuji, T. Suzuki,A. Kobayashi, J. Yokota, T. Sakiyama, T. Shibata, M. Yamamoto, S. Hirohashi,Loss of Keap1 function activates Nrf2 and provides advantages for lung cancercell growth, Cancer Res. 68 (2008) 1303–1309.

[58] T. Shibata, A. Kokubu, M. Gotoh, H. Ojima, T. Ohta, M. Yamamoto, S. Hirohashi,Genetic alteration of Keap1 confers constitutive Nrf2 activation and resistanceto chemotherapy in gallbladder cancer, Gastroenterology 135 (2008)1358–1368.

[59] T. Shibata, T. Ohta, K.I. Tong, A. Kokubu, R. Odogawa, K. Tsuta, H. Asamura,M. Yamamoto, S. Hirohashi, Cancer related mutations in Nrf2 impair its re-cognition by Keap1-Cul3 E3 ligase and promote malignancy, Proc. Natl. Acad.Sci. USA 105 (2008) 13568–13573.

[60] Y.R. Kim, J.E. Oh, M.S. Kim, M.R. Kang, S.W. Park, J.Y. Han, H.S. Eom, N.J. Yoo, S.H. Lee, Oncogenic Nrf2 mutations in squamous cell carcinomas of oesophagusand skin, J. Pathol. 220 (2010) 446–451.

[61] L.A. Muscarella, P. Parrella, V. D'Alessandro, A. la Torre, R. Barbano, A. Fontana,A. Tancredi, V. Guarnieri, T. Balsamo, M. Coco, M. Copetti, F. Pellegrini, P. DeBonis, M. Bisceglia, G. Scaramuzzi, E. Maiello, V.M. Valori, G. Merla,G. Vendemiale, V.M. Fazio, Frequent epigenetics inactivation of Keap1 gene innon-small cell lung cancer, Epigenetics 6 (2011) 710–719.

[62] J. Adam, E. Hatipoglu, L. O'Flaherty, N. Ternette, N. Sahgal, H. Lockstone,D. Baban, E. Nye, G.W. Stamp, K. Wolhuter, M. Stevens, R. Fischer, P. Carmeliet,P.H. Maxwell, C.W. Pugh, N. Frizzell, T. Soga, B.M. Kessler, M. El-Bahrawy, P.J. Ratcliffe, P.J. Pollard, Renal cyst formation in Fh1-deficient mice is in-dependent of the Hif/Phd pathway: roles for fumarate in Keap1 succinationand Nrf2 signaling, Cancer Cell 20 (2011) 524–537.

[63] A. Ooi, J.C. Wong, D. Petillo, D. Roossien, V. Perrier-Trudova, D. Whitten, B.W. Min, M.H. Tan, Z. Zhang, X.J. Yang, M. Zhou, B. Gardie, V. Molinié, S. Richard,P.H. Tan, B.T. Teh, K.A. Furge, An antioxidant response phenotype shared be-tween hereditary and sporadic type 2 papillary renal cell carcinoma, CancerCell 20 (2011) 511–523.

[64] G.M. DeNicola, F.A. Karreth, T.J. Humpton, A. Gopinathan, C. Wei, K. Frese,D. Mangal, K.H. Yu, C.J. Yeo, E.S. Calhoun, F. Scrimieri, J.M. Winter, R.H. Hruban,C. Iacobuzio-Donahue, S.E. Kern, I.A. Blair, D.A. Tuveson, Oncogene-inducedNrf2 transcription promotes ROS detoxification and tumorigenesis, Nature 475(2011) 106–109.

[65] S.L. Slocum, T.W. Kensler, Nrf2: control of sensitivity to carcinogens, Arch.Toxicol. 85 (2011) 273–284.

[66] Y. Mitsuishi, K. Taguchi, Y. Kawatani, T. Shibata, T. Nukiwa, H. Aburatani,M. Yamamoto, H. Motohashi, Nrf2 redirects glucose and glutamine into ana-bolic pathways in metabolic reprogramming, Cancer Cell 22 (2012) 66–79.

[67] D. Deeb, X. Gao, Y. Liu, D. Jiang, G.W. Divine, A.S. Arbab, S.A. Dulchavsky, S.C. Gautam, Synthetic triterpenoid CDDO prevents the progression and me-tastasis of prostate cancer in TRAMP mice by inhibiting survival signaling,Carcinogenesis 32 (2011) 757–764.

[68] X. Gao, D. Deeb, Y. Liu, A.S. Arbab, G.W. Divine, S.A. Dulchavsky, S.C. Gautam,Prevention of prostate cancer with oleanane synthetic triterpenoid CDDO-Mein the TRAMP mouse model of prostate cancer, Cancers, 3, Basel (2011), p.3353–3369.

[69] K. Tran, R. Risingsong, D. Royce, C.R. Williams, M.B. Sporn, K. Liby, The syn-thetic triterpenoid CDDO-methyl ester delays estrogen receptor-negativemammary carcinogenesis in polyoma middle T mice, Cancer Prev. Res. (Phi-ladelphia) 5 (2012) 726–734.

[70] D.S. Hong, R. Kurzrock, J.G. Supko, X. He, A. Naing, J. Wheler, D. Lawrence, J.P. Eder, C.J. Meyer, D.A. Ferguson, J. Mier, M. Konopleva, S. Konoplev,M. Andreeff, D. Kufe, H. Lazarus, G.I. Shapiro, B.J. Dezube, A phase I first-in-human trial of bardoxolone methyl in patients with advanced solid tumorsand lymphomas, Clin. Cancer Res. 18 (2012) 3396–3406.

[71] Y.Y. Wang, Y.X. Yang, H. Zhe, Z.X. He, S.F. Zhou, Bardoxolone methyl (CDDO-Me) as a therapeutic agent: an update on its pharmacokinetic and pharma-codynamic properties, Drug Des. Dev. Ther 8 (2014) 2075–2088.

[72] S. Nagaraj, J.I. Youn, H. Weber, C. Iclozan, L. Lu, M.J. Cotter, C. Meyer, C.R. Becerra, M. Fishman, S. Antonia, M.B. Sporn, K.T. Liby, B. Rawal, J.H. Lee, D.I. Gabrilovich, Anti-inflammatory triterpenoid blocks immune suppressivefunction of MDSCs and improves immune response in cancer, Clin. Cancer Res.16 (2010) 1812–1823.

[73] H. Satoh, T. Moriguchi, K. Taguchi, J. Takai, J.M. Maher, T. Suzuki, P.T. Winnard,V. Raman, M. Ebina, T. Nukiwa, M. Yamamoto, Nrf2-deficiency creates a re-sponsive microenvironment for metastasis to the lung, Carcinogenesis 31(2010) 1833–1843.

[74] K. Hiramoto, H. Satoh, T. Suzuki, T. Moriguchi, J. Pi, T. Shimosegawa,M. Yamamoto, Myeloid lineage-specific deletion of antioxidant system en-hances tumor metastasis, Cancer Prev. Res. (Philadelphia) 7 (2014) 835–844.

[75] P. Rada, A.I. Rojo, S. Chowdhry, M. McMahon, J.D. Hayes, A. Cuadrado, SCF/{beta}-TrCP promotes glycogen synthase kinase 3-dependent degradation ofthe Nrf2 transcription factor in a Keap1-independent manner, Mol. Cell. Biol.31 (2011) 1121–1133.

[76] S. Chowdhry, Y. Zhang, M. McMahon, C. Sutherland, A. Cuadrado, J.D. Hayes,Nrf2 is controlled by two distinct β-TrCP recognition motifs in its Neh6 do-main, one of which can be modulated by GSK-3 activity, Oncogene 32 (2013)

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–100 99

3765–3781.[77] K. Taguchi, I. Hirano, T. Itoh, M. Tanaka, A. Miyajima, A. Suzuki, H. Motohashi,

M. Yamamoto, Nrf2 enhances cholangiocyte expansion in Pten-deficient livers,Mol. Cell. Biol. 34 (2014) 900–913.

[78] K.N. Lewis, E. Wason, Y.H. Edrey, D.M. Kristan, E. Nevo, R. Buffenstein, Reg-ulation of Nrf2 signaling and longevity in naturally long-lived rodents, Proc.Natl. Acad. Sci. USA 112 (2015) 3722–3727.

[79] A.I. Rojo, P. Rada, M. Mendiola, A. Ortega-Molina, K. Wojdyla, A. Rogowska-Wrzesinska, D. Hardisson, M. Serrano, A. Cuadrado, The PTEN/Nrf2 axis pro-motes human carcinogenesis, Antioxid. Redox Signaling 21 (2014) 2498–2514.

[80] M. Komatsu, H. Kurokawa, S. Waguri, K. Taguchi, A. Kobayashi, Y. Ichimura, Y.S. Sou, I. Ueno, A. Sakamoto, K.I. Tong, M. Kim, Y. Nishito, S. Iemura,T. Natsume, T. Ueno, E. Kominami, H. Motohashi, K. Tanaka, M. Yamamoto, Theselective autophagy substrate p62 activates the stress responsive transcription

factor Nrf2 through inactivation of Keap1, Nat. Cell Biol. 12 (2010) 213–223.[81] Y. Ichimura, S. Waguri, Y.S. Sou, S. Kageyama, J. Hasegawa, R. Ishimura, T. Saito,

Y. Yang, T. Kouno, T. Fukutomi, T. Hoshii, A. Hirao, K. Takagi, T. Mizushima,H. Motohashi, M.S. Lee, T. Yoshimori, K. Tanaka, M. Yamamoto, M. Komatsu,Phosphorylation of p62 activates the Keap1-Nrf2 pathway during selectiveautophagy, Mol. Cell 51 (2013) 618–631.

[82] Y. Inami, S. Waguri, A. Sakamoto, T. Kouno, K. Nakada, O. Hino, S. Watanabe,J. Ando, M. Iwadate, M. Yamamoto, M.S. Lee, K. Tanaka, M. Komatsu, Persistentactivation of Nrf2 through p62 in hepatocellular carcinoma cells, J. Cell Biol.193 (2011) 275–284.

[83] K. Taguchi, N. Fujikawa, M. Komatsu, T. Ishii, M. Unno, T. Akaike, H. Motohashi,M. Yamamoto, Keap1 degradation by autophagy for the maintenance of redoxhomeostasis, Proc. Natl. Acad. Sci. USA 109 (2012) 13561–13566.

T. Suzuki, M. Yamamoto / Free Radical Biology and Medicine 88 (2015) 93–100100