Logo of biochemjBJ Latest papers and much more!
Biochem J. 2006 Nov 1; 399(Pt 3): 373–385.
Published online 2006 Oct 13. Prepublished online 2006 Jul 27. doi:  10.1042/BJ20060725
PMCID: PMC1615900

Negative regulation of the Nrf1 transcription factor by its N-terminal domain is independent of Keap1: Nrf1, but not Nrf2, is targeted to the endoplasmic reticulum


Nrf1 (nuclear factor-erythroid 2 p45 subunit-related factor 1) and Nrf2 regulate ARE (antioxidant response element)-driven genes. At its N-terminal end, Nrf1 contains 155 additional amino acids that are absent from Nrf2. This 155-amino-acid polypeptide includes the N-terminal domain (NTD, amino acids 1–124) and a region (amino acids 125–155) that is part of acidic domain 1 (amino acids 125–295). Within acidic domain 1, residues 156–242 share 43% identity with the Neh2 (Nrf2-ECH homology 2) degron of Nrf2 that serves to destabilize this latter transcription factor through an interaction with Keap1 (Kelch-like ECH-associated protein 1). We have examined the function of the 155-amino-acid N-terminal polypeptide in Nrf1, along with its adjacent Neh2-like subdomain. Activation of ARE-driven genes by Nrf1 was negatively controlled by the NTD (N-terminal domain) through its ability to direct Nrf1 to the endoplasmic reticulum. Ectopic expression of wild-type Nrf1 and mutants lacking either the NTD or portions of its Neh2-like subdomain into wild-type and mutant mouse embryonic fibroblasts indicated that Keap1 controls neither the activity of Nrf1 nor its subcellular distribution. Immunocytochemistry showed that whereas Nrf1 gave primarily cytoplasmic staining that was co-incident with that of an endoplasmic-reticulum marker, Nrf2 gave primarily nuclear staining. Attachment of the NTD from Nrf1 to the N-terminus of Nrf2 produced a fusion protein that was redirected from the nucleus to the endoplasmic reticulum. Although this NTD–Nrf2 fusion protein exhibited less transactivation activity than wild-type Nrf2, it was nevertheless still negatively regulated by Keap1. Thus Nrf1 and Nrf2 are targeted to different subcellular compartments and are negatively regulated by distinct mechanisms.

Keywords: antioxidant response element, endoplasmic reticulum, Kelch-like ECH-associated protein 1 (Keap1), nuclear factor-erythroid 2 p45 subunit-related factor 1 (Nrf1), nuclear factor-erythroid 2 p45 subunit-related factor 2 (Nrf2), oxidative stress
Abbreviations: ARE, antioxidant response element; bZIP, basic region-leucine zipper; β-gal, β-galactosidase; CNC, cap ‘n’ collar; CREB, cyclic-AMP-response-element-binding protein; DAPI, 4′,6-diamidino-2-phenylindole; FITC, fluorescein isothiocyanate; Gal4D, Gal4 DNA-binding domain; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GCLM, glutamate cysteine ligase modifier; Keap1, Kelch-like ECH-associated protein 1; KO, knockout; LDS, lithium dodecyl sulfate; Luc, luciferase; Maf, musculoaponeurotic fibrosarcoma retrovirus transforming gene; MEF, mouse embryonic fibroblast; Neh2L, Nrf2-ECH homology 2-like; NHB, N-terminal homology box; NQO1, NAD(P)H:quinone oxidoreductase 1; Nrf1, Nrf2, and Nrf3, nuclear factor-erythroid 2 p45 subunit-related factors 1, 2 and 3; NTD, N-terminal domain; NuPAGE, Novex® PAGE; P, promoter (as in P−1016/nqo1-Luc); PBGD, porphobilinogen deaminase; SREBP, sterol-regulatory-element-binding protein; SV40, simian virus 40; TCF11, transcription factor 11; TK, thymidine kinase promoter; UAS, a Gal4-specific binding site


Cells express many inducible genes that allow adaptation to oxidative stress. These include those encoding NQO1 [NAD(P)H:quinone oxidoreductase 1; nqo1], glutathione S-transferase, GCLM (glutamate cysteine ligase modifier; gclm) subunit, ferritin and haem oxygenase-1 [13]. Transcriptional activation of these genes occurs through AREs (antioxidant response elements) present in their promoter regions and is mediated by the CNC bZIP (cap ‘n’ collar basic-region leucine zipper) proteins Nrf1 (nuclear factor-erythroid 2 p45 subunit-related factor 1) and Nrf2 [4]. These two factors both bind AREs as heterodimers with small Maf (musculoaponeurotic fibrosarcoma retrovirus transforming gene) proteins [5,6]. Together, Nrf1 and Nrf2 regulate ARE-driven genes such as nqo1 and gclm [7,8].

Nrf1 and Nrf2 make distinct contributions to the antioxidant status of cells in vivo. Global KO (knockout) of Nrf1 has been shown to result in death of mice in utero [9,10]. Liver-specific KO of Nrf1 in mice at birth led to non-alcoholic steatohepatitis that developed into hepatoma [11]. Unlike disruption of nrf1, disruption of nrf2 yields mice that develop normally and do not spontaneously develop cancer [12], though they are more sensitive to chemical carcinogenesis [13,14]. Given the facts that Nrf1 and Nrf2 regulate common genes and they are both expressed in the liver [15], it is surprising that hepatic KO of these factors in mice yields different phenotypes. It therefore appears that, in vivo, Nrf2 is unable to compensate for the loss of Nrf1 in mouse liver, but the reason for this is not known.

Although it is axiomatic that the distinct biological activities of Nrf1 and Nrf2 are determined by their primary structures, it is unclear what specific features of the bZIP proteins are responsible for them producing markedly different phenotypes in the KO mice. Lack of detailed information about the structure and function of Nrf1 represents an obvious impediment to our understanding of the mechanism(s) by which it makes its unique contribution to maintenance of redox status. In this context, a complicating factor is that Nrf1 exists as multiple isoforms. In humans, a long version of Nrf1 has been described, designated TCF11 (transcription factor 11), which comprises 772 amino acids [16,17]. Human Nrf1 protein is slightly shorter than TCF11 and contains 742 amino acids; it arises by alternative splicing of exon 4 [1618]. Despite this difference, Nrf1 and TCF11 both transactivate ARE-driven reporter genes to a similar extent [19]. The mouse does not possess an orthologue of TCF11, rather it only expresses a full-length Nrf1 protein of 741 amino acids that shares 95% sequence identity with human Nrf1 [20].

The most obvious feature that distinguishes Nrf1 from Nrf2, and may therefore account for their distinct biological functions, is the presence of an additional 155-amino-acid N-terminal polypeptide in the former bZIP protein. This N-terminal extension in Nrf1 includes 31 amino acids that were considered to be part of the N-terminal acidic domain in TCF11 [16,17]. The function of neither the N-terminal 124 amino acids nor residues 125–155 in Nrf1 is known.

An additional explanation for the putative distinct functions of Nrf1 and Nrf2 is that they are differentially regulated by Keap1 (Kelch-like ECH-associated protein 1), a substrate adaptor for the Cullin-3:Rbx1 E3 ubiquitin ligase [21,22]. Keap1 controls the stability of Nrf2, ensuring that the bZIP protein is held at low levels under normal homoeostatic conditions, but can rapidly accumulate during oxidative stress when the substrate adaptor function of Keap1 is inhibited [23,24]. Nrf2 interacts with Keap1 through its N-terminal Neh2 (Nrf2-ECH homology 2) domain [25,26]. This interaction involves binding through both a high-affinity ETGE motif and a low-affinity DLG motif in the Neh2 domain of Nrf2 [27,28]. In Nrf1, the regions corresponding to the high- and low-affinity Keap1-binding sites in Nrf2 are conserved. Using a quantitative yeast two-hybrid assay, Kobayashi et al. [29] found that Keap1 can interact with Nrf1, albeit with only about 28% of the affinity Keap1 has for Nrf2. However, following transfection into mouse 3T3 cells, these workers also found that Keap1 did not influence transactivation of a reporter gene by Nrf1 [29]. It is therefore uncertain whether Keap1 regulates Nrf1.

In the present study we sought to identify: (i) the function of the additional 155-amino-acid N-terminal region in Nrf1 that is absent from Nrf2; (ii) whether Nrf1 is negatively regulated by Keap1 through its Neh2-like region located between amino acids 156–242.


Generation of Nrf1 expression constructs

The cDNA fragment coding full-length Nrf1 was obtained using the ProSTAR ultra HF RT (reverse-transcription)-PCR system (Stratagene, La Jolla, CA, U.S.A.) supplemented with mouse kidney RNA and primers Nrf1F and Nrf1R. The cDNA product was purified and subcloned into the KpnI/XbaI sites of pcDNA3.1/V5His B (Invitrogen) that translates into a C-terminally V5 (GKPIPNPLLGLDST)-tagged protein. Nrf2 expression constructs have been described previously [23]. The fidelity of the cDNA product and all constructs made in this study were sequence-verified by the Human Genome Group (Department of Molecular and Cellular Pathology, University of Dundee, Dundee, Scotland, U.K.). All oligonucleotide primers (see Supplementary Table 1 at http://www.BiochemJ.org/bj/399/bj3990373add.htm) were synthesized by MWG Biotech Co (Ebersberg, Germany).

Mutagenesis of Nrf1 and Keap1 constructs

Two strategies were employed for mutagenesis. First, relatively short internal deletion mutants of Nrf1, encoding Nrf1Δ125–170, Nrf1ΔETGE and Nrf1ΔDIDLID/DLG were produced using the Quick-Change Site Directed Mutagenesis kit (Stratagene) as described previously [30]; relevant pairs of sense and antisense primers are listed in Supplementary Table 1 at http://www.BiochemJ.org/bj/399/bj3990373add.htm. Secondly, the cDNA fragments encoding N- and C-terminally truncated Nrf1 proteins (Nrf1Δ2–120, Nrf1Δ2–150 and Nrf1Δ2–170) were generated by PCR through utilizing their corresponding primers (see Supplementary Table 1 at http://www.BiochemJ.org/bj/399/bj3990373add.htm); these amplified fragments were also subcloned into the KpnI/XbaI sites of pcDNA3.1/V5His B.

Constructs for Nrf1 fused with Gal4 or Nrf2

The pcDNA3.1Gal4D-V5 based expression constructs encoding the Gal4D (Gal4 DNA-binding domain) fused to various Nrf1 mutants were generated by first subcloning the Gal4D-encoding fragment of pGBT9 (Clontech Laboratories) into the HindIII site of a modified pcDNA3.1/V5His B plasmid. A total of 11 different-length cDNA fragments encoding different portions of Nrf1 were produced using PCR and equivalent GN primer pairs (see Supplementary Table 1 at http://www.BiochemJ.org/bj/399/bj3990373add.htm) and cloned into the BamHI/EcoRI sites of pcDNA3.1Gal4D-V5. The transactivation activity of these Gal4D fusion constructs was determined using the Gal4 luciferase reporter PTKUAS-Luc [31], where UAS is a Gal4-specific binding site. The Gal4D chimaeras containing the NTD (N-terminal domain) were subjected to immunocytochemistry. In order to obtain NTD–Nrf2 chimaeras, the XhoI/XbaI fragment of Nrf1/pcDNA3.1/V5His B was replaced by either the PCR product for Nrf2f (full-length mouse Nrf2 with Met1 and Met2 mutated into valine) or the PCR product for Nrf2s (residues 16–597 of Nrf2 with Met17 mutated into isoleucine); the resulting pcDNA3.1/V5His B expression constructs encoded chimaeric proteins containing 125 N-terminal amino acids from Nrf1 fused to either residues 1–597 of Nrf2 (giving NTD–Nrf2f) or to residues 16–597 of Nrf2 (giving NTD–Nrf2s).

Reporter constructs

The P 1016/nqo1-Luc and a mutant (mut1) containing a scrambled ARE across nucleotides −450 to −415 within the same construct have been described elsewhere [32]. Three different 41 bp ARE-driven luciferase reporter constructs (i.e. PSV40nqo1-ARE-Luc, PSV40gclm-ARE-Luc, and PSV40PBGD-ARE-Luc, where PBDG is porphobilinogen deaminase and SV40 is simian virus 40) were made by the insertion of the appropriately hybridized complementary oligonucleotides [3234] with 2 bp overhangs into the XhoI/BglII restriction sites of a modified pGL3-Promoter reporter vector. The pGL-6×ARE was a gift from Dr Jun Wang (Biomedical Research Centre, University of Dundee, Dundee, U.K.) and was engineered by inserting six copies of the oligonucleotide 5′-CCCGTGACAAAGCA-3′ into pGL3-Basic (Promega); the core ARE from the promoter of the rat glutathione S-transferase A2 gene is underlined [35]. The Gal4 luciferase reporter PTKUAS-Luc was a gift from Dr Mattias Gustafsson (Biomedical Research Centre, University of Dundee, Dundee, U.K.) and was made by cloning the SalI/XhoI fragment of pLacZr, containing a module of Gal4-specific binding sites, UAS×4, linked to the minimal TK (thymidine kinase promoter) of herpes simplex virus [31], into pGL3-Basic (Promega) cleaved with XhoI. The pcDNA3.1/V5His/lacZ, pcDNA4/HisMax/lacZ (Invitrogen) and pRL-TK (Promega) were used to control for transfection efficiency. On some occasions a PTKnqo1-ARE-Luc reporter plasmid was used to reduce background transactivation; this construct was made by replacing the BglII/HindIII fragment containing the SV40 promoter in PSV40nqo1-ARE-Luc reporter with an equivalent region encompassing the TK from the pRL-TK.

Cell culture, transfection and luciferase reporter assays

Rat liver RL-34 epithelial cells and monkey kidney COS-1 cells were grown as described previously [23,32]. MEFs (mouse embryonic fibroblasts) from wild-type (keap1+/+) and Keap1 KO (keap1−/−) mice [36] were grown in Iscove's Modified Dulbecco's medium (Invitrogen) supplemented with 10% (v/v) fetal calf serum, 10 μg/ml of insulin, 5.6 μg/ml transferrin, 6.7 ng/ml sodium selenite, 0.25% (w/v) NaHCO3, 100 units/ml penicillin and 100 μg/ml streptomycin at 37 °C in humidified air with 5% CO2. Equal numbers of cells (3×105 cells/well) were seeded and allowed to grow for 18–24 h in six-well plates. Once cells reached ∼60% confluence, they were transfected using Lipofectin® reagent or Lipofectamine™ 2000 (Invitrogen) according to the manufacturer's instructions. Approx. 36–48 h following transfection, gene reporter activity was determined using the Dual Luciferase Reporter Assay System (Promega). The ARE- and UAS-driven firefly luciferase activities were normalized to that of the Renilla reporter, or β-galactosidase (β-gal) reporter constructs used to control for transfection efficiency [23]. After subtraction of background reporter activity, obtained following co-transfection with ARE-null reporter (i.e., pGL3-Basic) and empty expression vectors, Nrf1-mediated activation activity was calculated. Significant differences in the transactivation activities were determined using the Student's t test.

Western blotting

Cells were disrupted in 1×RIPA buffer [50 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1% Ipegal (Nonidet P40), 0.5% sodium deoxycholate, 0.1% SDS, 1 mM dithiothreitol and a protease inhibitor cocktail]. The clarified cell lysates were examined using two electrophoretic systems. In the first method, samples were diluted with 5×SDS sample buffer, pH 6.8, and then heated at 100 °C for 10 min before being subjected to separation by SDS/PAGE in a Tris/glycine discontinuous system [37]. In the second method, cell lysates were diluted in 4×LDS (lithium dodecyl sulfate) sample buffer, pH 8.4, and heated at 70 °C for 10 min before being resolved by LDS/NuPAGE® (Novex, Invitrogen). Subsequently, Western-blot analysis was performed as reported previously [30]. C-terminally tagged Nrf1 and Nrf2 proteins were located with primary antibodies against the V5 epitope tag (Invitrogen) and visualized using peroxidase-labelled secondary antibody against mouse IgG. Immunoblotting using an anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase) antibody (Amino Bioproducts Ltd, Frellstedt, Germany) served as a loading control.

Immunocytochemistry and confocal microscopy

Cells (1×105) were grown on a coverslip in six-well plates. Approx. 24 h after transfection, they were fixed with 4% (v/v) paraformaldehyde in 20 mM sodium phosphate buffer, pH 6.8. The cells were washed four times, for 3 min each, in phosphate buffer before each immunocytochemistry step. Fixed cells were permeabilized by incubation in 0.2% Triton X-100 before non-specific antibody binding sites were blocked with 5% FBS (fetal bovine serum) and 0.5% normal goat serum. The coverslips were incubated with a primary antibody against V5 tag or Nrf1 for 1 h at 4 °C before being incubated with FITC for an additional 1 h at 4 °C. Nuclear DNA was stained in a solution containing 10 μg/ml of DAPI (4′,6-diamidino-2-phenylindole). The confocal images were obtained using an LSM 510 laser-scanning microscope system (Carl Zeiss, Jena, Germany). To address the question of whether Nrf1 can localize to the endoplasmic reticulum, the pDsRed2-ER (BD Biosciences Clontech, CA, U.S.A.) vector was employed that encodes a fusion of DsRed, a red fluorescent protein from the coral Discosoma sp., with the endoplasmic-reticulum targeting sequence of calreticulin coupled to its N-terminus and the endoplasmic-reticulum retention sequence tetrapeptide KDEL linked to its C-terminus.


Identification of domains in Nrf1

Bioinformatic analysis of Nrf1 suggested that it is a modular protein. As shown in Table 1, we have divided it into nine domains. Nrf1 is most obviously distinguished from Nrf2 by the presence of the NTD and an acidic/polar region (residues 125–155) located at its N-terminus. We included the latter region in acidic domain 1, as has been done previously during characterization of TCF11 [16,17]. Within the acidic domain 1, and immediately adjacent to the additional 155-amino-acid polypeptide in Nrf1, is a stretch of amino acid residues (156–242) that shares 43% identity with the N-terminal Neh2 region of Nrf2. Most significantly, this Neh2-like (Neh2L) subdomain in Nrf1 contains DLG and ETGE motifs that, in the case of Nrf2, are responsible for its interaction with Keap1 [23,27]. The present paper is concerned with the function of the NTD and the Neh2L subdomain in Nrf1.

Table 1
Protein domains in mouse Nrf1

Nrf1 is less potent than Nrf2 at transactivating gene expression

Although both Nrf1 and Nrf2 positively regulate ARE-driven gene expression [4], they may display selectivity in the genes they recognize. The AREs in the mouse nqo1 and gclm gene promoters were chosen for study because examination of nrf1 and nrf2 KO MEFs suggests nqo1 and gclm are targets for Nrf1 and Nrf2 [7]. Also, the ARE in human PBGD was studied because previously a random oligonucleotide-based screen to select transcription-factor binding sites showed its ARE to be optimal for Nrf1–MafG binding [38]. Figure 1(A) shows that Nrf1 activated a P−1016/nqo1-ARE-Luc reporter gene about 2.5-fold, but did not activate the P−1016/nqo1-mut1 reporter gene. By comparison, a parallel experiment revealed that Nrf2 transcriptionally activated the same mouse nqo1-ARE-Luc reporter gene about 5.0-fold.

Figure 1
Nrf1 is less potent than Nrf2 at transactivating ARE-driven reporter gene expression

On the basis of the mass of DNA transfected, Nrf1 exhibited less transactivation activity than Nrf2 towards three reporter genes driven by 41-bp ARE-containing sequences from the promoters of nqo1, gclm and PBGD (Figure 1B). To determine whether differences in protein level might account for the lower potency of Nrf1, immunoblotting experiments were undertaken. Surprisingly, although Nrf1 was less active than Nrf2, it clearly appeared to be the more abundant of the two proteins following transfection into COS-1 cells (Figure 1C). It was also noted that Nrf1 was resolved by SDS/PAGE into two or three major closely migrating polypeptides, whereas examination of Nrf2 in the same analysis gave a single Western-blot band. Presumably Nrf1 is subject to some type of post-translational modification that does not occur in the case of Nrf2. In addition to the major electrophoretic Nrf1 bands, three relatively minor smaller polypeptides were also observed that, on the basis of published work [1719,39], probably arise from translation being initiated from internal in-frame ATG codons.

The NTD negatively regulates transactivation by Nrf1

To explore why Nrf1 is less potent than Nrf2 at transactivating ARE-driven gene expression, various deletion mutants of Nrf1 were generated and tested in transfection experiments. The most striking difference we observed occurred following omission of the NTD from Nrf1. The resulting factor exhibited significantly greater transactivation activity than full-length Nrf1; this effect was more obvious in RL-34 cells than in COS-1 cells. As shown in Figure 2(A), removal of residues 2–120 from Nrf1 caused a substantial increase in transactivation by the truncated transcription factor in both cell lines. Further deletion of N-terminal amino acids up to residue 170 caused no additional increase in transactivation activity. It was consistently noted that, on the basis of mass of both expression vector transfected and protein levels expressed in cells (results not shown), Nrf1Δ2–120 affected levels of transactivation similar to those affected by Nrf2. This indicates that the NTD is probably the major negative regulating domain in Nrf1.

Figure 2
Nrf1 is negatively controlled by its NTD

To further test whether the NTD negatively controls Nrf1 activity, expression constructs were generated for the Gal4D protein fused to either full-length Nrf1 or one of a series of truncated proteins (Figure 2B, left upper panel). Transfection of COS-1 cells with constructs expressing the Gal4D protein fused at its C-terminus to a series of polypeptides generated from Nrf11–607, but lacking increasingly large portions from the N-terminus, showed that removal of residues 1–44 from Nrf1 had little effect on Gal4/UAS-driven reporter-gene expression (Figure 2B, cf. lanes 2 and 3). However, removal of residues 1–65 or 1–119 caused a gradual step-like increase in transactivation of Gal4 reporter activity from ∼600-fold to 1300-fold. Maximal activation occurred when almost the entire NTD was deleted (Figure 2B, lane 8). No obvious difference in the level of the different ectoptic fusion proteins was observed (Figure 2B, lower three panels).

The DIDLID/DLG and ETGE sequences in the Neh2L subdomain of Nrf1 neither influence its transactivation activity nor its cytoplasmic localization

Results presented in Figure 2 show that the NTD negatively regulates Nrf1. However, Nrf1 also contains, in close proximity to the NTD, a region that is similar to the Neh2 degron in Nrf2 (a degron is an element that is both necessary and sufficient to confer metabolic instability); in the case of Nrf2, the Neh2 region controls the half-life of the bZIP protein through an interaction with Keap1 [23]. If Nrf1 were controlled by Keap1, it would be predicted that deletion of the DIDLID/DLG sequence (amino acids 171–186) or ETGE motif (amino acids 234–237) in the Neh2L subdomain would increase transactivation by, and nuclear accumulation of, the mutant bZIP protein. To determine whether either these motifs in Neh2L or immediately adjacent N-terminal sequences control the function of Nrf1, we deleted either residues 125–170, the DIDLID/DLG sequence or the ETGE motif from an expression construct encoding the full-length bZIP protein (Figure 3A). Transfection experiments showed that removal of residues 125–170 in Nrf1 significantly decreased reporter-gene activity when compared with that obtained by wild-type Nrf1. In the same series of experiments, deletion of the DIDLID/DLG sequence caused a modest decrease in the reporter transactivation activity in COS-1 cells, but this was not observed in RL-34 cells (Figure 3B, upper panels). No significant difference in transactivation of reporter gene activity by Nrf1ΔETGE from that by wild-type Nrf1 was observed in either COS-1 or RL-34 cells. As neither an increase in reporter-gene activity nor a change in protein levels was observed in these Nrf1 mutants (Figure 3B), we conclude that neither the DIDLID/DLG sequence nor the ETGE motif negatively regulate the transcription factor. The inference from this finding is that it is unlikely Keap1 negatively controls Nrf1.

Figure 3
The ETGE and DIDLID/DLG motifs in Nrf1 neither regulate its activity nor control its subcellular localization

Confocal microscopy was performed to determine whether the DIDLID/DLG sequence and ETGE motif in Neh2L influence the subcellular localization of Nrf1. In COS-1 cells, ectopic wild-type Nrf1 gave stronger cytoplasmic staining than nuclear staining, whereas ectopic Nrf2 was located primarily in the nucleus (Figure 3C). Quantification of this staining pattern indicated that 70% of wild-type Nrf1 was located in the cytoplasm (Figure 3D). No differences were observed between the subcellular distributions of wild-type Nrf1 and its mutants Nrf1ΔDIDLID/DLG and Nrf1ΔETGE (Figures 3C and D). Identical results were obtained in RL-34 cells transfected with the same constructs (results not shown). These results indicate that the cytoplasmic–nuclear distribution of Nrf1 is unaffected by deletion of the DIDLID/DLG sequence and ETGE motif. Furthermore, the finding that Nrf1 is less potent than Nrf2 at transactivating ARE-driven reporter-gene expression may be explained by the observation that it is primarily located in the cytoplasm.

Keap1 does not antagonize Nrf1

In the case of Nrf2, the difference between transactivation of an ARE-reporter gene by the wild-type protein and the ΔETGE mutant is only obvious when the appropriate expression plasmids for the bZIP factor are co-transfected with an expression construct for Keap1 [23]. To test whether this also holds true for Nrf1, we co-transfected COS-1 cells with an ARE-driven reporter gene, along with expression constructs for Nrf1 and wild-type Keap1 (Figure 4A). Transactivation of ARE-driven luciferase activity by ectopic Nrf1 was unaffected by co-transfection with Keap1 (Figure 4A, left panel). By contrast, Keap1 reduced transactivation of reporter-gene activity by Nrf2 to approx. 50% of that observed by Nrf2 in the absence of ectopic Keap1. Further transfection experiments revealed that activation of a PTKnqo1-ARE-Luc reporter gene by ectopic Nrf1 was essentially identical in keap1+/+ and keap1−/− MEFs (Figure 4A, middle panel); the small increase of luciferase activity in the knockout MEFs may be due to the enhanced endogenous Nrf2 activity in these cells. By contrast, transfection of an expression construct for Nrf2 gave significantly higher reporter gene activity in keap1−/− MEFs than in keap1+/+ MEFs. We therefore conclude that full-length Nrf1 is not regulated by Keap1.

Figure 4
Keap1 does not negatively regulate Nrf1

To test the hypothesis that the Neh2L subdomain is capable of interacting with Keap1 but is prevented from doing so by the presence of the NTD, we transfected MEF cell lines with expression constructs for wild-type Nrf1 and Nrf1Δ2–120 along with a luciferase reporter plasmid. Consistent with previous data (Figure 2A), Nrf1Δ2–120 exhibited greater transactivation activity than did wild-type Nrf1 (Figure 4A, middle panel). Most importantly, no difference was observed in the transactivation of reporter gene activity by Nrf1Δ2–120 in keap1+/+ and keap1−/− MEFs (Figure 4A, middle panel). These data again suggest that Keap1 does not regulate Nrf1, even when the bZIP factor lacked its NTD. This conclusion was supported by the further observation that activation of pGL-6×ARE-Luc activity by wild-type Nrf1 or Nrf1Δ2–120 was not significantly affected by the reintroduction of Keap1 into keap1−/− MEFs in a rescue-type experiment (Figure 4A, right panel).

Ectopically expressed Nrf1 showed a similar subcellular distribution in both keap1+/+ and keap1−/− MEFs (Figure 4B). The predominantly cytoplasmic distribution of wild-type Nrf1 did not differ from that of Nrf1ΔDIDLID/DLG and Nrf1ΔETGE in keap1−/− MEFs.

The NTD is essential for cytoplasmic localization of Nrf1

Confocal microscopy was used to determine whether the NTD could control the extranuclear localization of Nrf1. Following transfection of mutant forms of Nrf1 into COS-1 and RL-34 cells, it was found that cytoplasmic location of Nrf1 was abolished upon removal of its NTD. The Nrf1Δ2–120 mutant protein was located primarily in the nucleus rather than cytoplasm (Figure 5A). Deletion of other single domains gave staining that was indistinguishable from wild-type Nrf1 (results not shown). These findings suggest that the NTD targets Nrf1 to a region within the cytoplasm.

Figure 5
The NTD of Nrf1 is required for its association with the endoplasmic reticulum

In transfection experiments, extranuclear ectopic Nrf1 co-localized with the ER–DsRed marker (Figures 5A and and5B).5B). This finding suggests that full-length Nrf1 is specifically targeted to this organelle. Overall, our data demonstrate that Nrf1 activity and its cytoplasmic localization are negatively regulated by the NTD, and that the negative effect of this domain on transactivation is independent of Keap1.

The NTD directs Nrf2 fusion proteins to the endoplasmic reticulum

To test whether the NTD is capable of targeting proteins to the endoplasmic reticulum, we generated expression constructs for proteins in which the NTD was attached to the N-terminus of wild-type Nrf2. Confocal microscopy showed that, in COS-1 and RL-34 cells, ectopically expressed wild-type Nrf2 was located primarily in the nucleus (Figures 3C and and5B).5B). Ectopic wild-type Nrf2 was also located in the nucleus of keap1−/− MEFs (Figure 6A). Conversely, a predominantly cytoplasmic localization was found following transfection of keap1−/− MEFs with expression constructs encoding fusion proteins in which the NTD from Nrf1 (residues 1–125) was attached to the N-terminus of either full-length Nrf2 (Nrf2f, residues 1–597) or short Nrf2 (Nrf2s, residues 16–597). The cytoplasmic staining of NTD–Nrf2f and NTD–Nrf2s was identical with that observed with ER–DsRed.

Figure 6
Attachment of the NTD from Nrf1 to Nrf2 results in negative regulation of the resulting fusion protein and their targeting to the endoplasmic reticulum

Luciferase reporter assays showed that, by comparison with wild-type Nrf2, NTD–Nrf2f and NTD–Nrf2s were less potent at transactivating an ARE-driven reporter gene (Figure 6B). Interestingly, a higher level of transactivation by the NTD–Nrf2 fusion proteins was observed in keap1−/− MEFs than in wild-type MEFs (Figure 6B), but this increase in activity was prevented by transfection into the mutant MEFs of an expression construct for Keap1 (Figure 6C). These results suggest that the Nrf2 chimaeras are negatively regulated by the NTD through a mechanism that involves tethering to the endoplasmic reticulum. Significantly, these NTD–Nrf2 fusion proteins were found to be regulated by Keap1, suggesting that the presence of the NTD does not obscure the effects of the Neh2 domain within the chimaera.

To determine whether the NTD can redirect the subcellular localization of not only Nrf2 but also an unrelated protein such as Gal4D, we transfected COS-1 cells with the Gal4D–Nrf1 fusion constructs. As shown in Figure 2(B), the fusion proteins from Gal4D–Nrf11–607 and Gal4D–Nrf11–298, which contained the NTD, were almost exclusively located in the endoplasmic reticulum; conversely the NTD-deficient fusion proteins from Gal4D–Nrf1120–607 and Gal4D–Nrf1120–298 were primarily expressed in the nucleus (Figure 6D); they had significant transactivation activity when compared with Gal4D–Nrf11–607. In addition, the Gal4D polypeptide from pcDNA3.1Gal4D-V5 was found to reside exclusively in the nucleus. No change in protein levels expressed by these constructs was observed (results not shown). These data indicate that an endoplasmic-reticulum-targeting signal exists within the NTD of Nrf1.


Gene KO experiments in the mouse have shown that Nrf1, but not Nrf2, is essential for embryonic development [7,9,10,40]. Furthermore, liver-specific KO of Nrf1 leads to hepatoma, whereas no such phenotype arises in nrf2−/− mice. The basis for the phenotypic differences in these mutant mice is not known. In an attempt to address this issue we have examined the function of the NTD in Nrf1 because it is absent from Nrf2. In addition, we have examined whether the Neh2L region in Nrf1 results in it being inhibited by Keap1, because the Neh2 domain in Nrf2 is responsible for this latter bZIP protein being degraded rapidly in a Keap1-dependent fashion under normal homoeostatic conditions.

Targeting of Nrf1 to the endoplasmic reticulum through its NTD

Our reporter gene experiments showed that deletion of the entire NTD from Nrf1 caused a substantial increase in ARE-driven transcription, but deletion of other domains in the bZIP protein did not result in similar activation. This indicates that the NTD is the principal negative mode by which the factor is regulated. We have also found that the negative regulation of Nrf1 is associated with an ability of NTD to direct the protein to the endoplasmic reticulum. Thus, whereas ectopic wild-type Nrf1 gave broad cytoplasmic staining when examined by confocal microscopy, Nrf1Δ2–120 gave almost exclusively nuclear staining when similarly studied. The conclusion that the NTD negatively controls Nrf1 by directing it to the endoplasmic reticulum is supported by our finding that NTD–Nrf2f, NTD–Nrf2s and NTD–Gal4D chimaeric proteins were all directed to this organelle, whereas wild-type Nrf2 and Gal4D were essentially exclusively nuclear proteins. Furthermore, NTD–Nrf2f and NTD–Nrf2s exhibited less transactivation activity towards the ARE-reporter gene than wild-type Nrf2, suggesting that, as expected, targeting these bZIP proteins to membranes inhibits their activity.

Bioinformatic analyses suggested that the N-terminal part of the NTD is responsible for Nrf1 being directed to the endoplasmic reticulum. Proteins, such as calreticulin, calnexin and calmegin, which are known to be located in the endoplasmic reticulum, each contain at their N-terminus a region with significant sequence homology with Nrf1. In particular, the peptide sequence GLLQFTILLSLIGVRVD found between residues 11 and 27 of Nrf1 shares 35% identity with the MLLSVPLLLGLLGLAAA leader sequence found between residues 1 and 17 of mouse calreticulin [41]. The latter sequence has been implicated in the translocation of calreticulin into the endoplasmic reticulum, suggesting that the GLLQFTILLSLIGVRVD peptide in Nrf1 is probably responsible for it being targeted to the same organelle.

Besides Nrf1, a small number of other transcription factors have been shown to be targeted to the endoplasmic reticulum. These include the sterol-regulatory-element-binding proteins SREBP1 and SREBP2 [42], as well as ATF6 (activating transcription factor 6) and the related cyclic-AMP-response-element-binding proteins CREB4 and CREBH [4345]. Transcription factors that are located in the endoplasmic reticulum all possess at least one transmembrane domain that anchors them to the organelle in such a fashion that their DNA-binding and transactivation domains are orientated towards the cytosolic face of the membrane. This topology is necessary to ensure that when such factors are released from intracellular membranes they are free to translocate to the nucleus. Release of SREBP and CREBH from the endoplasmic reticulum entails trafficing to the Golgi apparatus and regulated intramembrane proteolysis, a process that involves cleavage of the factors in response to specific stimuli by the sequential actions of Site-1 and Site-2 proteases. It is not know whether Nrf1 is processed by these proteases, but bioinformatic analysis predicts that residues 81–90 form an amphipathic α-helix and within this region there exists the tetrapeptide RRLL that could be a recognition sequence for the Site-1 peptidase [4648]. In addition, on the basis of comparison with proteins containing tripartite signal sequences, it is predicted that flanking the hydrophobic sequence between residues 11 and 24 in Nrf1 there exists a signal peptidase cleavage site at the C-terminal side of Thr30 [49]. The presence of proteolytic cleavage sites in the NTD requires further examination, but their existence seems highly probable given the heterogeneity in electrophoretic mobility of Nrf1 observed in Figure 1.

Conserved motifs in the NTD of Nrf1 and Nrf3

An unexpected finding during the present study was that the NTD in Nrf1 appears to be represented in Nrf3. Comparison between the N-terminal regions of Nrf1 and Nrf3 revealed two conserved regions, GLLQXTILLSLXGXRVDXD (where X indicates any amino acid) and RLLXXVRALXXXXXPXTXVXAWLVH, found between residues 11 and 29, and 82 and 106, respectively in Nrf1, and between residues 12 and 30, and 76 and 100, in Nrf3. We refer to these as the NHB1 (N-terminal homology box 1) and NHB2, with the former residing closest to the N-termini of the factors. As noted above, NHB1 is predicted to be part of an endoplasmic-reticulum-targeting sequence on the basis of its homology with the N-termini of calreticulin, and NHB2 is predicted to form part of an amphipathic α-helix that may be cleaved by Site-1 protease.

The presence of NHB1 and NHB2 in Nrf3 suggests that it may also be associated with the endoplasmic reticulum. This hypothesis requires to be tested. The remaining two Nrf family members, Nrf2 and NF-E2 p45, appear unlikely to locate to the endoplasmic reticulum because they do not possess sequences related to either NHB1 or NHB2. In fact, as has been demonstrated in Figures 3, ,55 and and66 of the present paper, Nrf2 resides primarily in the nucleus. Similarly, NF-E2 p45 also locates primarily to the nucleus [50].

Keap1 does not regulate Nrf1

The Nrf2 transcription factor is negatively regulated by Keap1 through at least two interactions that involve, on the one hand, the low-affinity DLG motif and the high-affinity ETGE motif in its Neh2 domain, and, on the other hand, the Kelch-repeat domain in Keap1 [23,27,28,5154]. These interactions are redox-dependent and result in the stability of Nrf2 being controlled by Keap1 under normal homoeostatic conditions [3,26,53]. In the Neh2L subdomain of Nrf1, the DLG motif and the ETGE motif are both conserved [52]. Before the present study was undertaken, it was unclear from the literature whether this bZIP protein was regulated by Keap1; a previous two-hybrid analysis had shown that Nrf1 can interact with Keap1 in yeast cells, but that Keap1 does not inhibit Nrf1 transactivation activity in mouse 3T3 cells [29]. Our data show that deletion of the DLG and ETGE motifs in Nrf1 does not influence either its transactivation activity or its cytoplasmic localization. We have found that, in wild-type and mutant MEFs, Keap1 does not influence the subcellular distribution of Nrf1, nor does it alter the ability of the bZIP protein to activate ARE-driven gene expression.

In view of the fact that Nrf1 is subject to N-terminal proteolysis, we investigated the possibility that the NTD might somehow mask Keap1-binding sites in the Neh2L subdomain; we postulated that proteolytic removal of the NTD at the endoplasmic reticulum might alleviate some sort of steric hindrance, thereby allowing the Neh2L subdomain to interact with Keap1. However, we obtained no experimental data to support this hypothesis. On transfection into keap1+/+ and keap1−/− MEFs, we found that transactivation of ARE-driven gene expression by Nrf1Δ2–120 was not influenced by the presence of Keap1. During these experiments we did, however, show that Keap1 can negatively regulate NTD–Nrf2 chimaeric proteins. In this case, attachment of either the NTD to Nrf2f (i.e. full-length Nrf2) or the N-terminal 125 amino acids of Nrf1 to Nrf2s (i.e. residues 16–597 of Nrf2) resulted in decreased transactivation activity, but it did not obscure the inhibitory effect of Keap1 on transactivation of an ARE-driven luciferase reporter gene. These data indicate that differences between the effect of Keap1 on Nrf1 and Nrf2 cannot be attributed to the presence of the NTD.

Our experiments involving transfection into keap1−/− MEFs of the Nrf1 NTD deletion mutants show that the presence of Neh2L in Nrf1 is insufficient to result in its inhibition by Keap1. The following three reasons may account for failure of Keap1 to inhibit Nrf1: (i) differences around the ETGE motif in Nrf1 may reduce its affinity for Keap1; (ii) differences around the DLG motif in Nrf1 may compromise the interaction with Keap1; and (iii) loss of critical ubiquitin acceptor lysine residues between the DLG and ETGE motifs in Nrf1 may block ubiquitylation. Given the report that, in yeast two-hybrid experiments, Nrf1 possesses only about 28% of the affinity for Keap1 than Nrf2 it exhibits for Keap1 [29], it appears most likely that changes around the ETGE motif in the Neh2L subdomain of Nrf1 are responsible for the apparent reduced interaction between Nrf1 and Keap1. It has recently been reported that the LDEETGEFL nonapeptide between residues 76 and 84 of Nrf2 is involved in the high-affinity interaction of the bZIP protein with Keap1 [55]. As this sequence is represented by VDGETGESF (residues 231–239) in Nrf1, it is clear that there are four amino acid changes around the ETGE motif (i.e. Val231, Gly233, Ser238 and Phe239) that could be responsible for the inability of Keap1 to inhibit the bZIP factor. It is not known which, if any, of these four residues might be responsible for the reduced affinity between Nrf1 and Keap1. It is also possible that changes around the DLG motif in Nrf1 inactivated the low-affinity Keap1-binding site. In this case, the functional sequence in Nrf2 is QDIDLGVSR (residues 26–34), whereas that in Nrf1 is QDIDLGAGR (residues 181–189); thus the presence of Ala187 and/or Gly188 in Nrf1 may be responsible for the loss of Keap1 inhibition. Lastly, it could be argued that the reason Keap1 does not inhibit Nrf1 is because the lysine residues to which ubiquitin is conjugated in Nrf2 are not conserved in Nrf1. In the case of Nrf2, the lysine residues that are targeted for ubiquitin conjugation by cullin3:Rbx1 (i.e. Lys44, Lys50, Lys52, Lys53, Lys56, Lys64 and Lys68) are located in an α-helical region, residues 39–71, held between the DLG and ETGE motifs [22,56,57]. Sequence alignment shows that, among the seven lysine residues in Nrf2 that can accept ubiquitin, only two are conserved in Nrf1 (i.e. Lys199 and Lys205). It is not known whether this is insufficient to allow ubiquitination of Nrf1. It is also possible that the protein fold within the Neh2L region prevents Lys199 and Lys205 from being available for ubiquitination. These possibilities should be explored in future.

Concluding comments

In the present study we have discovered that transactivation by Nrf1 is negatively regulated by its NTD through a mechanism that entails an association of the factor with the endoplasmic reticulum. We have provided evidence that the negative regulation of Nrf1-mediated transactivation and control of its cytoplasmic location are independent of Keap1. Our findings that Nrf1 can associate with the endoplasmic reticulum and is not regulated by Keap1, whereas Nrf2 does not associate with this organelle and is regulated by Keap1, suggest that these two bZIP factors mediate adaptation to redox stress in different subcellular compartments. Presumably this will explain, at least in part, the distinct phenotypes observed in the Nrf1 and Nrf2 KO mice.

While our work was under revision, a paper [58] was published that reports, like ours, that Nrf1 is targeted through its NTD to the endoplasmic reticulum. In addition to these observations, we have demonstrated that Keap1 does not negatively regulate Nrf1.

Online data

Supplementary Table 1 :


This work was supported by both the Association for International Cancer Research (grant 03-074) and Tenovus Scotland (grant T05/6). We thank Dr Michael McMahon, Dr Larry G. Higgins, Dr Tim W. P. Devling and Dr Paul Nioi (all of the Biomedical Research Centre, Ninewells Hospital and Medical School, University of Dundee, Dundee, Scotland, U.K.) for providing expression constructs and for critical advice. We are indebted to Dr M. Gustafsson and Dr J. Wang for gifts of reporter constructs.


1. Hayes J. D., Ellis E. M., Neal G. E., Harrison D. J., Manson M. M. Cellular response to cancer chemopreventive agents: contribution of the antioxidant responsive element to the adaptive response to oxidative and chemical stress. Biochem. Soc. Symp. 1999;64:141–168. [PubMed]
2. Nguyen T., Sherratt P. J., Pickett C. B. Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu. Rev. Pharmacol. Toxicol. 2003;43:233–260. [PubMed]
3. Nguyen T., Sherratt P. J., Nioi P., Yang C. S., Pickett C. B. Nrf2 controls constitutive and inducible expression of ARE-driven genes through a dynamic pathway involving nucleocytoplasmic shuttling by Keap1. J. Biol. Chem. 2005;280:32485–32492. [PubMed]
4. Venugopal R., Jaiswal A. K. Nrf1 and Nrf2 positively and c-Fos and Fra1 negatively regulate the human antioxidant response element-mediated expression of NAD(P)H:quinone oxidoreductase1 gene. Proc. Natl. Acad. Sci. U.S.A. 1996;93:14960–14965. [PMC free article] [PubMed]
5. Motohashi H., O'Connor T., Katsuoka F., Engel J. D., Yamamoto M. Integration and diversity of the regulatory network composed of Maf and CNC families of transcription factors. Gene. 2002;294:1–12. [PubMed]
6. Venugopal R., Jaiswal A. K. Nrf2 and Nrf1 in association with Jun proteins regulate antioxidant response element-mediated expression and coordinated induction of genes encoding detoxifying enzymes. Oncogene. 1998;17:3145–3156. [PubMed]
7. Leung L., Kwong M., Hou S., Lee C., Chan J. Y. Deficiency of the Nrf1 and Nrf2 transcription factors results in early embryonic lethality and severe oxidative stress. J. Biol. Chem. 2003;278:48021–48029. [PubMed]
8. Chen L., Kwong M., Lu R., Ginzinger D., Lee C., Leung L., Chan J. Y. Nrf1 is critical for redox balance and survival of liver cells during development. Mol. Cell. Biol. 2003;23:4673–4686. [PMC free article] [PubMed]
9. Farmer S. C., Sun C. W., Winnier G. E., Hogan B. L., Townes T. M. The bZIP transcription factor LCR-F1 is essential for mesoderm formation in mouse development. Genes Dev. 1997;11:786–798. [PubMed]
10. Chan J. Y., Kwong M., Lu R., Chang J., Wang B., Yen T. S., Kan Y. W. Targeted disruption of the ubiquitous CNC-bZIP transcription factor, Nrf-1, results in anemia and embryonic lethality in mice. EMBO J. 1998;17:1779–1787. [PMC free article] [PubMed]
11. Xu Z., Chen L., Leung L., Yen T. S., Lee C., Chan J. Y. Liver-specific inactivation of the Nrf1 gene in adult mouse leads to nonalcoholic steatohepatitis and hepatic neoplasia. Proc. Natl. Acad. Sci. U.S.A. 2005;102:4120–4125. [PMC free article] [PubMed]
12. Chan K., Lu R., Chang J. C., Kan Y. W. NRF2, a member of the NFE2 family of transcription factors, is not essential for murine erythropoiesis, growth, and development. Proc. Natl. Acad. Sci. U.S.A. 1996;93:13943–13948. [PMC free article] [PubMed]
13. Ramos-Gomez M., Kwak M. K., Dolan P. M., Itoh K., Yamamoto M., Talalay P., Kensler T. W. Sensitivity to carcinogenesis is increased and chemoprotective efficacy of enzyme inducers is lost in nrf2 transcription factor-deficient mice. Proc. Natl. Acad. Sci. U.S.A. 2001;98:3410–3415. [PMC free article] [PubMed]
14. Iida K., Itoh K., Kumagai Y., Oyasu R., Hattori K., Kawai K., Shimazui T., Akaza H., Yamamoto M. Nrf2 is essential for the chemopreventive efficacy of oltipraz against urinary bladder carcinogenesis. Cancer Res. 2004;64:6424–6431. [PubMed]
15. McMahon M., Itoh K., Yamamoto M., Chanas S. A., Henderson C. J., McLellan L. I., Wolf C. R., Cavin C., Hayes J. D. The cap'n'collar basic leucine zipper transcription factor Nrf2 (NF-E2 p45-related factor 2) controls both constitutive and inducible expression of intestinal detoxification and glutathione biosynthetic enzymes. Cancer Res. 2001;61:3299–3307. [PubMed]
16. Luna L., Skammelsrud N., Johnsen O., Abel K. J., Weber B. L., Prydz H., Kolsto A. B. Structural organization and mapping of the human TCF11 gene. Genomics. 1995;27:237–244. [PubMed]
17. Luna L., Johnsen O., Skartlien A. H., Pedeutour F., Turc-Carel C., Prydz H., Kolsto A. B. Molecular cloning of a putative novel human bZIP transcription factor on chromosome 17q22. Genomics. 1994;22:553–562. [PubMed]
18. Chan J. Y., Han X. L., Kan Y. W. Cloning of Nrf1, an NF-E2-related transcription factor, by genetic selection in yeast. Proc. Natl. Acad. Sci. U.S.A. 1993;90:11371–11375. [PMC free article] [PubMed]
19. Husberg C., Murphy P., Martin E., Kolsto A. B. Two domains of the human bZIP transcription factor TCF11 are necessary for transactivation. J. Biol. Chem. 2001;276:17641–17652. [PubMed]
20. McKie J., Johnstone K., Mattei M. G., Scambler P. Cloning and mapping of murine Nfe2l1. Genomics. 1995;25:716–719. [PubMed]
21. Furukawa M., Xiong Y. BTB protein Keap1 targets antioxidant transcription factor Nrf2 for ubiquitination by the Cullin 3-Roc1 ligase. Mol. Cell. Biol. 2005;25:162–171. [PMC free article] [PubMed]
22. Zhang D. D., Lo S. C., Cross J. V., Templeton D. J., Hannink M. Keap1 is a redox-regulated substrate adaptor protein for a Cul3-dependent ubiquitin ligase complex. Mol. Cell. Biol. 2004;24:10941–10953. [PMC free article] [PubMed]
23. McMahon M., Itoh K., Yamamoto M., Hayes J. D. Keap1-dependent proteasomal degradation of transcription factor Nrf2 contributes to the negative regulation of antioxidant response element-driven gene expression. J. Biol. Chem. 2003;278:21592–21600. [PubMed]
24. Cullinan S. B., Gordan J. D., Jin J., Harper J. W., Diehl J. A. The Keap1-BTB protein is an adaptor that bridges Nrf2 to a Cul3-based E3 ligase: oxidative stress sensing by a Cul3–Keap1 ligase. Mol. Cell. Biol. 2004;24:8477–8486. [PMC free article] [PubMed]
25. Dinkova-Kostova A. T., Holtzclaw W. D., Kensler T. W. The role of Keap1 in cellular protective responses. Chem. Res. Toxicol. 2005;18:1779–1791. [PubMed]
26. Itoh K., Wakabayashi N., Katoh Y., Ishii T., Igarashi K., Engel J. D., Yamamoto M. Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain. Genes Dev. 1999;13:76–86. [PMC free article] [PubMed]
27. McMahon M., Thomas N., Itoh K., Yomamoto M., Hayes J. D. Dimerisation of substrate adaptors can facilitate cullin-mediated ubiquitylation of proteins by a ‘tethering’ mechanism: A two-site interaction model for the Nrf2–Keap1 complex. J. Biol. Chem. 2006;281:24756–24768. [PubMed]
28. Katoh Y., Iida K., Kang M. I., Kobayashi A., Mizukami M., Tong K. I., McMahon M., Hayes J. D., Itoh K., Yamamoto M. Evolutionary conserved N-terminal domain of Nrf2 is essential for the Keap1-mediated degradation of the protein by proteasome. Arch. Biochem. Biophys. 2005;433:342–350. [PubMed]
29. Kobayashi A., Ohta T., Yamamoto M. Unique function of the Nrf2–Keap1 pathway in the inducible expression of antioxidant and detoxifying enzymes. Methods Enzymol. 2004;378:273–286. [PubMed]
30. Zhang Y., Cho Y. Y., Petersen B. L., Bode A. M., Zhu F., Dong Z. Ataxia telangiectasia mutated proteins, MAPKs, and RSK2 are involved in the phosphorylation of STAT3. J. Biol. Chem. 2003;278:12650–12659. [PubMed]
31. Solomin L., Johansson C. B., Zetterstrom R. H., Bissonnette R. P., Heyman R. A., Olson L., Lendahl U., Frisen J., Perlmann T. Retinoid-X receptor signalling in the developing spinal cord. Nature. 1998;395:398–402. [PubMed]
32. Nioi P., McMahon M., Itoh K., Yamamoto M., Hayes J. D. Identification of a novel Nrf2-regulated antioxidant response element (ARE) in the mouse NAD(P)H:quinone oxidoreductase 1 gene: reassessment of the ARE consensus sequence. Biochem. J. 2003;374:337–348. [PMC free article] [PubMed]
33. Hudson F. N., Kavanagh T. J. Cloning and characterization of the proximal promoter region of the mouse glutamate–L-cysteine ligase regulatory subunit gene. Biochim. Biophys. Acta. 2000;1492:447–451. [PubMed]
34. Mignotte V., Wall L., deBoer E., Grosveld F., Romeo P. H. Two tissue-specific factors bind the erythroid promoter of the human porphobilinogen deaminase gene. Nucleic Acids Res. 1989;17:37–54. [PMC free article] [PubMed]
35. Rushmore T. H., Morton M. R., Pickett C. B. The antioxidant responsive element. Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J. Biol. Chem. 1991;266:11632–11639. [PubMed]
36. Wakabayashi N., Itoh K., Wakabayashi J., Motohashi H., Noda S., Takahashi S., Imakado S., Kotsuji T., Otsuka F., Roop D. R., Harada T., et al. Keap1-null mutation leads to postnatal lethality due to constitutive Nrf2 activation. Nat. Genet. 2003;35:238–245. [PubMed]
37. Laemmli U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–685. [PubMed]
38. Johnsen O., Murphy P., Prydz H., Kolsto A. B. Interaction of the CNC-bZIP factor TCF11/LCR-F1/Nrf1 with MafG: binding-site selection and regulation of transcription. Nucleic Acids Res. 1998;26:512–520. [PMC free article] [PubMed]
39. Kozak M. Pushing the limits of the scanning mechanism for initiation of translation. Gene. 2002;299:1–34. [PubMed]
40. Murphy P., Kolsto A. Expression of the bZIP transcription factor TCF11 and its potential dimerization partners during development. Mech. Dev. 2000;97:141–148. [PubMed]
41. Smith M. J., Koch G. L. Multiple zones in the sequence of calreticulin (CRP55, calregulin, HACBP), a major calcium binding ER/SR protein. EMBO J. 1989;8:3581–3586. [PMC free article] [PubMed]
42. Brown M. S., Goldstein J. L. The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a membrane-bound transcription factor. Cell. 1997;89:331–340. [PubMed]
43. Haze K., Yoshida H., Yanagi H., Yura T., Mori K. Mammalian transcription factor ATF6 is synthesized as a transmembrane protein and activated by proteolysis in response to endoplasmic reticulum stress. Mol. Biol. Cell. 1999;10:3787–3799. [PMC free article] [PubMed]
44. Stirling J., O'Hare P. CREB4, a transmembrane bZip transcription factor and potential new substrate for regulation and cleavage by S1P. Mol. Biol. Cell. 2006;17:413–426. [PMC free article] [PubMed]
45. Zhang K., Shen X., Wu J., Sakaki K., Saunders T., Rutkowski D. T., Back S. H., Kaufman R. J. Endoplasmic reticulum stress activates cleavage of CREBH to induce a systemic inflammatory response. Cell. 2006;124:587–599. [PubMed]
46. Ye J., Rawson R. B., Komuro R., Chen X., Dave U. P., Prywes R., Brown M. S., Goldstein J. L. ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Mol. Cell. 2000;6:1355–1364. [PubMed]
47. Cheng D., Espenshade P. J., Slaughter C. A., Jaen J. C., Brown M. S., Goldstein J. L. Secreted site-1 protease cleaves peptides corresponding to luminal loop of sterol regulatory element-binding proteins. J. Biol. Chem. 1999;274:22805–22812. [PubMed]
48. Toure B. B., Munzer J. S., Basak A., Benjannet S., Rochemont J., Lazure C., Chretien M., Seidah N. G. Biosynthesis and enzymatic characterization of human SKI-1/S1P and the processing of its inhibitory prosegment. J. Biol. Chem. 2000;275:2349–2358. [PubMed]
49. Martoglio B., Dobberstein B. Signal sequences: more than just greasy peptides. Trends Cell. Biol. 1998;8:410–415. [PubMed]
50. Husberg C., Murphy P., Bjorgo E., Kalland K. H., Kolsto A. B. Cellular localisation and nuclear export of the human bZIP transcription factor TCF11. Biochim. Biophys. Acta. 2003;1640:143–151. [PubMed]
51. McMahon M., Thomas N., Itoh K., Yamamoto M., Hayes J. D. Redox-regulated turnover of Nrf2 is determined by at least two separate protein domains, the redox-sensitive Neh2 degron and the redox-insensitive Neh6 degron. J. Biol. Chem. 2004;279:31556–31567. [PubMed]
52. Kobayashi M., Itoh K., Suzuki T., Osanai H., Nishikawa K., Katoh Y., Takagi Y., Yamamoto M. Identification of the interactive interface and phylogenic conservation of the Nrf2–Keap1 system. Genes Cells. 2002;7:807–820. [PubMed]
53. Velichkova M., Hasson T. Keap1 regulates the oxidation-sensitive shuttling of Nrf2 into and out of the nucleus via a Crm1-dependent nuclear export mechanism. Mol. Cell. Biol. 2005;25:4501–4513. [PMC free article] [PubMed]
54. Motohashi H., Yamamoto M. Nrf2-Keap1 defines a physiologically important stress response mechanism. Trends Mol. Med. 2004;10:549–557. [PubMed]
55. Padmanabhan B., Tong K. I., Ohta T., Nakamura Y., Scharlock M., Ohtsuji M., Kang M. I., Kobayashi A., Yokoyama S., Yamamoto M. Structural basis for defects of Keap1 activity provoked by its point mutations in lung cancer. Mol. Cell. 2006;21:689–700. [PubMed]
56. Zhang D. D., Lo S.-C., Cross J. V., Templeton D. J., Hannink M. Keap1 is a redox-regulated substrate adaptor protein for a Cul3-dependent ubiquitin ligase complex. Mol. Cell. Biol. 2004;24:10941–10953. [PMC free article] [PubMed]
57. Tong K. I., Katoh Y., Kusunoki H., Itoh K., Tanaka T., Yamamoto M. Keap1 recruits Neh2 through binding to ETGE and DLG motifs: characterization of the two-site molecular recognition model. Mol. Cell. Biol. 2006;26:2887–2900. [PMC free article] [PubMed]
58. Wang W., Chan J. Y. Nrf1 is targeted to the endoplasmic reticulum by an N-terminal transmembrane domain. Inhibition of nuclear translocation and transacting function. J. Biol. Chem. 2006;281:19676–19687. [PubMed]

Articles from Biochemical Journal are provided here courtesy of The Biochemical Society
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.
  • Taxonomy
    Taxonomy records associated with the current articles through taxonomic information on related molecular database records (Nucleotide, Protein, Gene, SNP, Structure).
  • Taxonomy Tree
    Taxonomy Tree

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...