Keap1 Cystenine 151 as a Potential Target for Artemisitene-Induced Nrf2 Activation

Artemisitene (ATT) activates the nuclear factor (erythroid-derived 2)-like 2 (Nrf2) by increasing its stabilization and reducing ubiquitination. The cysteine (Cys) residues of the cytosolic Nrf2 repressor Kelch-like ECH-associated protein-1 (Keap1) function as redox sensors and may be crucial in activating Nrf2. To determine whether ATT-induced Nrf2 activation is dependent on the modification of Keap1 and to elucidate the underlying mechanism, we transfected cell lines with six different Keap1 mutant constructs, each with a Cys (−77, −151, −257, −273, −288, and −297) to Ser substitution. Only the Cys151Ser mutant prevented ATT-mediated activation of Nrf2, indicating that the Cys151 residue of Keap1 likely interacts with ATT and is essential for Nrf2 stabilization and transcription of downstream genes. Our finding provides a pharmacological basis for using artemisitene against oxidative stress-related diseases.


Introduction
Nuclear factor (erythroid-derived 2)-like 2 (Nrf2) is a transcription factor that activates multiple genes, including drug transporters, antioxidant enzymes, and phase II detoxifying enzymes like heme oxygenase (HO-1) [1,2]. e Nrf2 signaling pathway is tightly regulated [3][4][5][6][7] by Keap1, the substrate adaptor protein of the cul3-dependent E3 ubiquitin ligase, which specifically binds to Nrf2 and results in the latter's polyubiquitination and cytoplasmic retention [8][9][10][11]. Keap1-dependent ubiquitination of Nrf2 is inhibited by oxidative stress, as well as the chemical inducers of Nrf2, which activates the Nrf2-dependent downstream protective genes [9]. Studies show that modification of specific cysteine (Cys) residues in Keap1 plays a critical role in the oxidative stress or chemicalinduced activation of Nrf2 [4,12]. Most chemical inducers of Nrf2 covalently modify the Cys151 residue of Keap1, which is also the target of reactive oxygen species (ROS) and other electrophiles. Both covalent and oxidative changes in Cys151 destabilize the Keap1-Cul3 interaction and subsequently activate the Nrf2-dependent downstream genes [13]. Site-directed mutagenesis on the conserved Cys residues of Keap1 by later studies showed that Cys77, Cys151, Cys257, Cys273, Cys288, and Cys293 residues are also critical for Nrf2 activation [9,[12][13][14][15]. Several phytochemicals have recently been shown to activate Nrf2. Artemisitene (ATT), a semisynthetic derivative of the sesquiterpene artemisinin isolated from Artermisia annua [1,5], can activate Nrf2 by blocking its ubiquitination and increasing its stability [16]. However, the underlying molecular mechanism is still unclear. In the present study, we found that ATT activated the Nrf2-dependent pathway by covalently modifying the Cys151 of Keap1, which provides a strong pharmacological basis for its future applications in oxidative stress-related diseases.

Cell Culture and Reagents.
Cos-1, A549, and 293T cell lines were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). All lines were checked for mycoplasma contamination at least once a month using the mycoplasma PCR detection kit. e cells were cultured in 10% fetal bovine serum-(FBS-) supplemented DMEM at 37°C under 5% CO 2 . Tert-butylhydroquinone (tBHQ) was purchased from Sigma Chemical (St. Louis, MO, USA) and ATT from Tianjin Silan Technology Co. Ltd. e cells from the 2 nd to 5 th passages were used for the assays.

Western Blotting.
e cells were washed with cold PBS and lysed on ice with NP-40 cell-lysis buffer supplemented with 2% 2-mercaptoethanol, 50 mM DTT, and 1% Protease Inhibitor Cocktail. e lysates were cleared by centrifuging for 15 minutes at 13,000 rpm, and the protein content of the supernatants was evaluated using the BCA assay. Equal amount of proteins per sample were denatured in the sample loading buffer by boiling for 5 min, resolved in 7.5% and 10% SDS-polyacrylamide gels, and then transferred onto polyvinylidene difluoride (PVDF) membranes. e latter was blocked with 5% skimmed milk in TBST (TRIS-buffered saline with 0.1% Tween-20) at room temperature for one hour and incubated overnight at 4°C with the primary antibodies against Nrf2 (1 : 1000; Abcam, Cambridge, UK, ab76026, Rabbit monoclonal), Keap1 (1 : 500; Proteintech, IL, USA, 10503-2-AP, Rabbit monoclonal), GAPDH (1 : 500; Cell Signaling, OH, USA), and HO-1 (1 : 1000; Abcam, Cambridge, UK, ab13243, Rabbit monoclonal). e blots were rinsed thrice with TBST and incubated with the horseradish peroxidase-conjugated secondary antibodies at room temperature for 1 hour. After washing thrice with TBST, the bands were developed using an ECL substrate solution (Super Signal ™ West Dura Extended Duration Substrate, ermo fisher) for 1 min [17]. e grey value of proteins was measured by ImageJ (NIH, Bethesda, MD, USA). Averages of three independent experiments were presented as the final data [18].

Transfection.
e cells were seeded in 6-well plates and transfected with the requisite plasmids diluted 1 : 1 in Lipofectamine ® 3000 ( ermo Fisher Scientific Life Sciences, Waltham, MA, USA) according to the manufacturer's instructions [19].

Luciferase Reporter Gene Assay.
Cos-1 cells were cotransfected with 40 ng each of the ARE-luciferase and Renilla luciferase expression plasmids, 80 ng of the wild or mutant-type Keap1 plasmid, and 80 ng Nrf2 plasmid using lipofectamine 3000. Forty-eight hours after transfection, the cells were treated with 7 μM ATT for 24 h. ARE-luciferase activity was quantified using the dual luciferase reporter assay system (Promega, CA, USA) and normalized to that of Renilla luciferase.

Detection of Intracellular ROS Levels.
Intracellular ROS levels were detected using the oxidation-sensitive fluorescent probe DCFH-DA. After treatment with varying doses (1 μM, 4 μM, and 7 μM) ATT or 0.05% DMSO with or without 10 mM ROSUP for 6 h or 16 h, the cells were washed twice with phosphate-buffered saline (PBS). After incubating with 10 μM DCFH-DA at 37°C for 20 min, the fluorescence intensity of the oxidized 2,7-dichlorofluorescein (DCF) was detected on a flow cytometer (Becton Dickinson). For each sample, 10000 events were analyzed.

ATT Activates the Nrf2 Signaling Pathway in a Keap1-Dependent Manner.
To determine whether ATT activates Nrf2 in a Keap1-dependent manner, we treated MB231 cells and A549 cells that harbor a mutated form of Keap1, with ATT. While no changes were seen in the Nrf2 levels in A549 cells after ATT treatment (Figures 1(a) and 1(b)), ATT upregulated Nrf2 in the MB231 cells (Figures 1(c) and 1(d)). However, knocking down Keap1 in the MB231 cells abrogated the effect of ATT (Figures 1(c) and 1(d)), which clearly indicated that ATT-induced Nrf2 activation is dependent on Keap1.

ATT Targets Keap1 at Cys151.
e cysteine residues in Keap1 act as redox sensors, and most known Nrf2 inducers such as tBHQ can induce covalent modification of Cys151 or other cysteine residues of Keap1 [3,13]. To determine whether ATT has a similar mechanistic basis, we constructed several Keap1 mutants, each with a single Cys to Ser mutation at residues 77, 151, 257, 273, 288, or 297. Cos-1 cells were transfected with wild-type or mutated keap1 constructs and treated with varying doses of ATT. A 68 KD band corresponding to the WT Keap1 protein was observed in the   untreated control lysates, and a larger 120 KD band was also observed in the ATT-treated lysates (Figure 2(a)), the intensity of which was dependent on the ATT dose (Figures 2(a)-2(c)). ATT treatment induced the larger band in all Keap1 mutant (77Cys-Ser, 257Cys-Ser, 273Cys-Ser, 288cys-Ser, and 297Cys-Ser) lysates, except the 151Cys-Ser mutant. Taken together, Keap1 is covalently modified by ATT at the Cys151 site (Figure 2(d)).

Keap1-Cys151 Is Essential for Nrf2 Activation by ATT.
To determine whether Keap1-Cys151 is essential for Nrf2 activation, the Cos-1 cells were cotransfected with WT/ C151S Keap1, Nrf2, and Renilla and ARE-luciferase plasmids. As expected, ATT significantly increased the AREdependent transcriptional activity in WT-Keap1 transfected cells compared to the untreated control. However, ATT had no effect on ARE-dependent transcriptional activity in the cells expressing Keap1-C151S (Figure 3(a)), which also showed decreased Nrf2-dependent transcriptional activity compared to cells expressing the WT Keap1 (Figure 3(a)). is indicated a loss of activation phenotype conferred by the C151S mutation and that Keap1-Cys151 is essential for Nrf2 activation by ATT. Furthermore, we explored whether Keap1-Cys151 is also essential for ATT-mediated Nrf2 stabilization. e Keap1-C151S cells expressed significantly lower Nrf2 levels after ATT treatment compared to the WT cells. Accordingly, ATT did not activate the Nrf2-dependent downstream gene HO-1 in the cells harboring the Keap1-C151S mutant (Figures 3(b)-3(d)). Taken together, Keap1-Cys151 is necessary for ATT-dependent Nrf2 stabilization and downstream transcriptional activation ( Figure 4).

ATT Activates the Nrf2
Pathway at Low Dose. Nrf2 levels were significantly increased following different chemical inducers compared to that in the untreated sample. However, while tBHQ and sulforaphane were only effective at high doses, ATT increased Nrf2 levels at low doses ranging from 1 to 5 μM. In addition, 3 μM ATT was nearly as effective as 10 μM SF and 100 μM tBHQ, indicating greater sensitivity of ATT. Figure 5, ATT treatment did not have any significant effect on the fluorescence intensity of the oxidized DCF compared to the untreated controls. us, ATT does not increase ROS production and cannot active Nrf2 by inducing oxidative stress.

Discussion
Nrf2 activation by specific chemicals prevents or inhibits the progression of oxidative stress-related disorders, including diabetes, cancer, cardiovascular diseases, neurodegenerative diseases, chronic obstructive pulmonary disease, pulmonary fibrosis, and inflammation in experimental animal models. ATT, a derivative of the antimalarial agent artemisinin, was recently identified as a novel Nrf2 activator [16]. In this study, we further elucidated the mechanisms underlying ATT-mediated activation of Nrf2 and its downstream pathway, which provide a crucial pharmacological basis for its future clinical applications.
Nrf2 is repressed by the Keap1 protein under physiological conditions, which is reversed upon exposure to reactive chemicals, oxidative stress, or chemopreventive agents. Zhang and Hannink first reported that Keap1 targets Nrf2 for ubiquitination and proteasome-dependent degradation through its Cys residues, and C151 in particular is required for stabilizing Nrf2 in response to tBHQ-and sulforaphane-induced oxidative stress [12]. e Cys residues in Keap1, especially C151 and C288, act as redox sensors for activation of Nrf2 [13] by undergoing covalent modification. In this study, we confirmed that covalent modification of the Cys151 residue is essential for Nrf2 activation by ATT. Based on the structural comparison between artemisitene and artemisinin, we concluded that the distinct ene bond in the former is critical for Cys151-dependent Nrf2 activation by ATT [16]. Taken together, an ATT-based therapeutic approach targeting Nrf2 is a promising strategy to control diseases associated with oxidative stress.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.