Triterpenoid electrophiles (avicins) activate the innate stress response by redox regulation of a gene battery

Avicins are proapoptotic and anti-inflammatory triterpene electrophiles isolated from an Australian desert tree, Acacia victoriae. The presence of two α,β unsaturated carbonyl groups (Michael reaction sites) in the side chain of the avicin molecule prompted us to study its effects on NF-E2–related factor 2 (Nrf2), a redox-regulated transcription factor that controls the expression of a battery of detoxification and antioxidant proteins via its binding to antioxidant response element (ARE). Avicin D–treated Hep G2 cells showed translocation of Nrf2 into the nucleus and a time-dependent increase in ARE activity. These properties were sensitive to DTT, suggesting that avicins affect one or more critical cysteine residues, probably on the Keap1 molecule. Downstream of ARE, an activation of a battery of stress-induced proteins occurred. The implications of these findings were evaluated in vivo in mouse skin exposed to an ancient stressor, UV light. Avicins inhibited epidermal hyperplasia, reduced p53 mutation, enhanced apoptosis, decreased generation of 8-hydroxy-2′-deoxyguanosine, and enhanced expression of NADPH:quinone oxidoreductase 1 and heme oxygenase-1. These data, combined with our earlier published work, demonstrate that avicins represent a new class of plant stress metabolites capable of activating stress adaptation and suppressing proinflammatory components of the innate immune system in human cells by redox regulation. The relevance for treatment of clinical diseases in which stress responses are dysfunctional or deficient is discussed.

Many chronic human diseases associated with aging, such as cancer, are characterized by either an inadequate response or a hyperresponse to chronic stress. It is therefore important to search for pharmacological ways to regulate the stress response in clinical medicine. We recently discovered a family of triterpenoid electrophiles, called avicins, that contain an unusual side chain that contains two Michael reaction sites (1). Avicins have been shown to induce apoptosis by direct perturbation of the mitochondria (1) and to inhibit the PI3K/Akt-signaling pathway (2). Avicin’s ability to suppress chemical-induced carcinogenesis by its anti-inflammatory, antimutagenic, and antioxidant properties (3) was supported by its inhibitory effect on NF-κB activation (4). The ability of DTT to block avicins’ suppression of NF-κB activation suggested that these compounds might be regulating critical redox-sensitive thiol groups. We therefore asked whether avicins could regulate other redox-dependent transcription factors involved in stress responses.

We chose to study the module involving NF-E2–related factor 2 (Nrf2) and antioxidant response element (ARE), which is responsible for the transcriptional activation of numerous detoxifying enzymes and antioxidant proteins. One of the properties common to inducers of phase 2 enzymes is their ability to modify sulfhydryl groups by oxidation, reduction, or alkylation (5). The structure of avicins (1) includes α,β unsaturated carbonyl groups (two in the outer monoterpene chains and one at C28) that could react with nucleophiles, especially cysteine sulfhydryl groups, by a Michael-type addition.

In the present study, we show that avicins activate a battery of stress-induced proteins via the Nrf2 pathway in an epithelial cell line derived from a hepatic carcinoma called Hep G2. The implications of these findings were evaluated in vivo in mouse skin exposed to UV light. The protective effects of avicins on UVB-induced skin damage are reported here. In this paper, we suggest that avicins represent a new class of electrophilic metabolites capable of significantly enhancing innate stress response, an ancient pathway to ensure organismal homeostasis against biotic and abiotic threats.

In vitro studies

Regulation of transcription. Nrf2, a member of the cap ’n’ collar family of bZip transcription factors, is normally repressed through its cytoplasmic localization by binding to the cytoskeleton-associated protein Keap1 (13). Exposure of cells to inducers of phase 2 enzymes disrupts the Keap1-Nrf2 complex, allowing Nrf2 to translocate into the nucleus, where it binds to the ARE and activates transcription. Fluorescence microscopy of Hep G2 cells transfected with pEGFP-Nrf2 revealed the nuclear localization of Nrf2 in response to avicin D treatment (Figure 1a). Western blot analysis of the nuclear fractions of avicin D–treated cells showed a gradual increase in Nrf2 levels with time, while they declined concomitantly in the cytoplasm (Figure 1b). Avicin D–induced changes in the nuclear and cytoplasmic Nrf2 levels were comparable to those induced by tert-butyl hydroquinone (t-BHQ), an antioxidant known to induce phase 2 proteins such as NQO1 via activation of ARE (ref. 14; see Supplemental Figure 2a, http://www.jci.org/cgi/content/full/113/1/65/DC1, for loading control and molecular weight markers).

Figure 1

Nuclear localization of Nrf2 and activation of ARE-mediated gene expression. (a) Hep G2 cells transfected with EGFP-Nrf2 were either untreated or treated with avicin D (2 μg/ml) for 1 hour at 37°C. Nuclear localization of Nrf2 was studied by fluorescence microscopy (H&E staining; ×400). (b) Hep G2 cells were treated with avicin D (2 μg/ml) for 0–4 hours at 37°C. Western blot analysis of cytoplasmic and nucleaE-LUC plasmid, and LUC activity was measured as described in Methods. Dose response of ARE activation was studied using 0–2 μg/ml of avicin D for 2 hours at 37°C (c), and kinetics of ARE activation was studied using 2 μg/ml of avicin D for 0–16 hours at 37°C (d). (e) Effect of DTT (100 μM) on ARE activation was studied by application of DTT to cells for 2 hours, before, after, or along with avicin D (2 μg/ml).

ARE, a cis-acting regulatory sequence identified in the promoter regions of several genes encoding phase 2 detoxification enzymes, binds to Nrf2 via a core sequence, 5′-TGACNNNGC-3′. Hep G2 cells transfected with hARE-LUC plasmid were exposed to avicin D, and changes in LUC activity were used as a measure of ARE activation. Figure 1c shows a dose-dependent increase of ARE-driven LUC activity with 1–2 μg/ml of avicin D inducing maximal activation. Avicin D also induced a time-dependent increase in LUC activity, starting at 30 minutes after treatment, peaking at 2 hours, and then gradually declining (Figure 1d).

The Keap1-Nrf2 complex is believed to serve as a cytoplasmic sensor of oxidative stress (13) based on the presence of highly reactive sulfhydryl groups in both these proteins (13, 15). To evaluate the role of cysteine residues in the action of avicins, we studied the effect of DTT on avicin D–induced ARE activation (Figure 1e). While DTT by itself induced a small increase in ARE activation, pretreatment of cells with DTT almost completely blocked the effect of avicin D. In contrast, DTT given after an avicin D treatment had minimal effect, suggesting the irreversibility of avicin D’s action. Incubating avicin D with DTT before putting it on cells had no effect on the activation, indicating that DTT does not interact with avicin D but probably competes with it for binding to cysteines that are critical for ARE activation. A recent report from Talalay’s group has identified four cysteine residues that are key sensors regulating induction of phase 2 enzymes in the Keap1 molecule (16). Whether avicin D interacts with one or more of these cysteines will be the subject of our future studies.

Activation of a gene battery. Following its effects on Nrf2 and ARE, we next looked at the effects of avicin D on a battery of downstream gene products. NQO1, a flavoprotein induced in response to a variety of biotic and abiotic stresses, utilizes NADPH as an electron donor to catalyze the two-electron reduction and detoxification of quinones and their derivatives. Figure 2a reveals a dose-dependent increase of NQO1 expression in Hep G2 cells treated with avicin D for 2 hours. Induction of NQO1 achieved with 1–2 μg/ml (0.5–1 μM) of avicin D was comparable to that achieved with 100 μM of t-BHQ (2-hour treatment) (see Supplemental Figure 2b, http://www.jci.org/cgi/content/full/113/1/65/DC1, for loading control and molecular weight markers). To correlate expression with activity, we next assayed for dicoumarol-sensitive NQO1 activity in avicin D–treated Hep G2 cells using dichloroindophenol as an electron acceptor. Study of the kinetics of enzyme activity revealed an optimal induction of NQO1 activity between 2 and 4 hours of treatment with avicin D (Figure 2b). A dose-dependent increase in NQO1 activity was seen following a 2-hour treatment with avicin D. The increase in activity (2.6-fold) observed with 2 μg/ml of avicin D (Figure 2c) correlated well with the increase in expression (2.7-fold) seen under similar conditions (Figure 2a).

Figure 2

Effect of avicin D on NQO1 and HO-1 enzymes. (a) Expression of NQO-1 was studied in Hep G2 cells treated with 0–2 μg/ml of avicin D for 2 hours, by Western blot analysis. (b and c) Dicoumarol-sensitive NQO1 activity was assayed in Hep G2 cells treated with 0–2 μg/ml of avicin D for 2 hours (b) and 2 μg/ml of avicin D for 0–4 hours (c), as described in Methods. NQO1 activity was expressed in nanomoles of dichlorophenolindohenol (DCPIP) reduced per minute per milligram protein. (d) Hep G2 cells treated with avicin D (2 μg/ml) for 0–16 hours were analyzed for changes in levels of HO-1 by Western blot analysis. (e) Hep G2 cells treated with avicin D (2 μg/ml) for 0–24 hours were assayed for bilirubin levels as described in Methods. (f) Synthesis of ferritin was studied in Hep G2 cells treated with avicin D (2 μg/ml) for 16–24 hours.

HO breaks down heme into equimolar amounts of biliverdin, carbon monoxide, and iron, all of which can be toxic. Mice lacking HO-1 are unable to modulate body iron stores and are susceptible to hepatic injury by iron (17). To study the effect of avicin D on HO-1, we first measured protein expression by Western blot analysis. A marked increase in HO-1 levels was observed at 8 and 16 hours after treatment (Figure 2d) with return to base line after 24 hours with avicin D (data not shown). The reason for the delayed induction of HO-1 (compared with the other ARE-regulated enzymes studied) is not clear. Besides ARE, the promoter region of the HO-1 gene contains consensus binding sites for NF-κB, activated protein-1, activated protein-2, IL-6 response element, and other transcription factors (18). Avicin-induced HO-1 levels could be dependent on the net effects of some or most of these transcription factors. It is also possible that the delay is at the step of protein synthesis and not transcription. Based on some reports that in stressed cells HO-1 is induced after the cellular thiols like GSH have been oxidized (19), it is possible that the delayed induction of HO-1 in avicin-treated cells could be a second phase of protection against oxidative stress. To determine whether increased HO-1 expression resulted in increased enzyme activity, we studied two of the downstream products of HO-1 activity, namely bilirubin and ferritin.

Bilirubin, the most abundant endogenous antioxidant in mammalian tissues, especially in human serum, is a potent scavenger of superoxide and peroxyl radicals (20). Therefore, it is exciting to observe that treatment of Hep G2 cells with avicin D resulted in a dramatic time-dependent increase in bilirubin levels (Figure 2e), which correlates with the striking increase in HO-1 activity. The induction of bilirubin could account for the antioxidant properties of avicins in lipid compartments of the cell.

Ferritin, an iron-storage protein, sequesters free iron and prevents its participation in reactive oxygen species (ROS) generation and thereby maintains cellular iron homeostasis. Oxidant stress activates multiple ferritin-regulation pathways (21), one of which is an ARE-mediated mechanism (22). As shown in Figure 2f, a twofold increase in ferritin synthesis was seen in Hep G2 cells following a 16-hour treatment with avicin D, which is consistent with the time of HO-1 activation. Thus, regulation of ferritin by avicins could contribute to their anti-inflammatory action.

The intracellular redox state is regulated by the contribution of different molecules, such as TRX, GSH/GSSG), and NADPH. High ratios of reduced to oxidized forms of TRX and GSH are maintained by the activities of TRX_red and GSH reductase, respectively. Figure 3a shows the effect of avicin D on auranofin-sensitive TRX_red activity, which increases within 30 minutes of treatment. GSH plays a key role in regulating intracellular redox as well as in detoxifying the cell of various potentially harmful molecules. It acts as a cofactor for GPx, a selenium-dependent enzyme known to be the predominant mechanism for reduction of H₂O₂ to water and of lipid peroxides to alcohol, and known to detoxify peroxynitrite. GSH renders a variety of electrophilic compounds less toxic by forming conjugates with them. These reactions require the action of GSTs, a ubiquitous family of detoxification enzymes. Figure 3b shows a time-dependent increase in the levels of GSH in avicin D–treated Hep G2 cells, which begin to decline after 3 hours of treatment. The modest increase and then decline in GSH levels could be accounted for by the constant utilization of GSH within the cell, as well as the possibility that avicin can bind to the cysteine in GSH, thereby quenching it. This is consistent with the increased activities of GPx and GST seen in these cells following treatment with avicin D (Figure 3, c and d). In confirmation of the increase in GPx and GST activities, we observed a decrease in lipid peroxidation both in avicin-treated cells and in a cell-free system (Supplemental Figure 3, http://www.jci.org/cgi/content/full/113/1/65/DC1). Cells in tissue culture are believed to be under a much higher oxidative stress than cells in vivo (23). The ability of avicins to reduce levels of cellular H₂O₂ as well as lipid peroxides under these conditions might explain the short half-life of these detoxification proteins.

Figure 3

Effect of avicin D on TRX_red (a), GSH (b), GPx (c), and GST levels (d). Hep G2 cells were treated with avicin D (2 μg/ml) for 0–300 minutes, and cells were lysed by the freeze-thaw method. Levels or activity of the different proteins were measured as described in Methods. The results shown for each of the proteins are representative of three different assays.

In vivo studies

Epidermal thickness. Mice exposed to UVB for 10 weeks show increased levels of hyperplasia and hyperkeratosis, with focal patches of dysplasia (Figure 4a). As shown in Figure 4b, pretreatment of skin with AE resulted in a significant decrease in hyperplasia. Application of AE after UVB irradiation was less effective in reducing hyperplasia (data not shown). These results were confirmed by BrdU staining (Supplemental Table 1, http://www.jci.org/cgi/content/full/113/1/65/DC1). SKH-1 mice treated for 2 weeks with AE alone showed no changes in epidermal thickness, cell proliferation, and apoptosis. This confirms our earlier observation in SENCAR mice, in which a 4-week treatment with AE alone had no effect on epidermal thickness or dermal cellularity (3).

Figure 4

Representative microphotographs showing the effect of AE on epidermal hyperplasia (a and b; H&E staining; ×200), p53 mutations (c and d; H&E staining; ×200), and apoptosis (e and f; H&E staining; ×400) in SKH-1 hairless mouse skin. Mice treated with acetone (a, c, and e) or 0.4 mg of AE (b, d, and f) were irradiated with UVB as described in Methods.

p53 mutations. The p53 protein plays a crucial role in the cellular response to UVB-induced DNA damage. The basal and suprabasal layers of the epidermis from mice that had been exposed to UVB for 10 weeks showed intense p53 immunoreactivity (Figure 4c) in the form of microscopic clusters of preneoplastic cells overexpressing mutant p53. Pretreatment with AE resulted in a significant decrease in the number of mutant p53–positive cells in these clusters (Figure 4d and Supplemental Table 1, http://www.jci.org/cgi/content/full/113/1/65/DC1). These results suggest that AE could possibly reverse the detrimental process of skin carcinogenesis by inhibiting UV-induced p53 mutations by one or more of the following mechanisms: (a) removal of damaged cells, (b) suppression of oxidative stress, and (c) repair of DNA damage.

Apoptosis. Based on the decrease in p53 mutations, we evaluated the effect of AE on apoptosis of skin cells. Compared with the untreated, UVB-exposed skin (Figure 4e), skin pretreated with AE prior to UVB irradiation showed an increased number of apoptotic cells (Figure 4f and Supplemental Table 1, http://www.jci.org/cgi/content/full/113/1/65/DC1).

Oxidative damage and immunohistochemistry of phase 2 enzymes. Free radicals are known to cause DNA damage via strand breakage and/or formation of modified bases like 8-OH-dG. UVB treatment led to an approximately fourfold increase in the ratio of 8-OH-dG per 100,000 pmol dG (10⁵dG) (pmol/pmol) in DNA of control mice (Figure 5). Application of AE to the skins immediately after UVB exposure or before UV exposure also decreased the 8-OH-dG levels significantly (Figure 5a).

Figure 5

(a) Effect of AE on 8-OH-dG formation. Skin sections from mice treated with AE after or before exposure to UVB were analyzed for 8-OH-dG levels as described in Methods. (b) Effect of AE on expression of NQO1 in mouse skin. Mice treated with acetone (control) or AE (0.4 mg) were exposed to UV. Immunohistochemistry of NQO1 (H&E staining; ×400) was done as described in Methods.

Based on the in vitro results from Hep G2 cells, we studied the effect of AE on NQO1 expression in vivo. AE-treated skin showed a twofold increase in NQO1 staining (Figure 5b). While exposure to UV increased the NQO1 expression by about 8%, pretreatment with AE prior to UV exposure resulted in a 15.7% increase in NQO1 staining (Figure 5b). Similar effects were seen on HO-1 expression in these skin samples (data not shown). These results indicate that the induction of phase 2 detoxification proteins by AE could contribute to its protective effects seen in vivo.

In her Nobel lecture two decades ago, Barbara McClintock discussed the innate ability of genomes to restructure when faced with external or internal threats (stress) (24). As organisms age, the ability to withstand chronic cellular stress diminishes, affecting virtually all organ systems, leading to a myriad of human diseases including cancer. Recent work has linked chronic cell injury and cellular repair with inadequate stress response, telomere shortening, gene mutations, genomic instability, and finally cancer (25). Of potential relevance is the evidence that stem cells demonstrate features of cells under stress (26). Identification of molecules that can prevent dysfunctioning of the redox machinery or repair the damage caused by it is therefore becoming important in clinical medicine (27). Transcription factors that regulate expression of antioxidant proteins and enzymes that regenerate them (28) form one such class of molecules. Often these sensor proteins contain critical cysteines (thiol groups) that modify the location of signaling molecules or transcription factors (28). Thus the importance of redox regulation of cysteines in oxidative and nitrosative stress is being increasingly recognized in clinical medicine (28).

In this paper we have described the ability of avicins to regulate the Nrf2-ARE system and induce a battery of genes that restore redox balance within the cell. The two electrophilic Michael addition sites (α,β unsaturated carbonyl groups) on the side chain of the avicin molecules are probably responsible for its binding to one or more critical cysteines, most likely on the Keap1 molecule, to allow for Nrf2 to translocate to the nucleus and activate the ARE. This speculation is consistent with the ability of avicins to activate OxyR (unpublished data), a bacterial transcription factor whose activity is regulated by cysteines (29) and has functions similar to those of Nrf2. Mass spectrometric studies with avicins and target proteins are underway to define the precise chemical changes. Alternative non–redox-dependent mechanisms of activation of Nrf2 (30) are also being studied.

The enhancement of the Nrf2-regulated gene battery could have important clinical ramifications for various forms of cancer and a multitude of other diseases that are characterized by an inadequate stress response (31). For example, individuals with polymorphism in the NQO1 gene have been shown to be at an increased risk of developing leukemias and other cancers (32). Bilirubin is known to be protective against physiological jaundice and oxygen radical–mediated injury (33). A recently published case study of a child deficient in HO (34) is illustrative of its role as an anti-inflammatory stress protein. HO-1 has also been implicated in a variety of cytoprotective roles, including protection against liver and pulmonary injury, acetaminophen toxicity, complement-dependent inflammatory response, and elimination of haem (35). Studies using knockout mice as well as mice overexpressing GPx have confirmed its antioxidant role in vivo (36). Decrease in activity of GST due to hypermethylation has been implicated in several malignancies, particularly prostate cancer (37). Interestingly, in postmitotic cells, enhancement of glial Nrf2 protects neurons from oxidative stress (38).

The ability of avicins to induce higher levels of both HO-1 and bilirubin, compared with GSH could be a novel and important finding. Excess GSH can suppress the stress response, in particular HO-1 (39). Thus the delayed and greater induction of HO-1 and downstream bilirubin probably tilts the balance in favor of anti-inflammation and immune suppression. This notion is supported by avicin’s ability to suppress the proinflammatory transcription factor NF-κB. Other compounds with α,β unsaturated carbonyl groups (40) show similar effects on the NF-κB and Nrf2 systems. Thus, a new concept is emerging that indicates to us that compounds with such signature components (α,β unsaturated carbonyl groups), by upregulating the innate stress response and downregulating components of the proinflammatory innate immune response (2–4), can reduce inflammation and oxidative damage (see model, Figure 6). These compounds also induce apoptosis of damaged cells by mechanisms that are closely associated with, but independent of, thiol regulation (refs. 1, 41; and our unpublished observations). This makes them potential therapeutic agents for reducing many forms of cellular stress.

Figure 6

A proposed model of avicin-induced enhancement of the innate stress response. By redox regulation of thiols, avicins have opposing effects on NF-κB–mediated and Nrf2-mediated transcription, resulting in downregulation of proinflammatory components of the innate immune response and upregulation of the intrinsic stress response. The early and late induction of Nrf2-regulated proteins enables a prolonged protective effect. Avicins’ proapoptotic response to stress, reported earlier, is functionally separate but linked to the enhancement of innate stress response. 4-HNE, 4-hydroxy-2-nonenal. OH•, hydroxyl radical.

Stress responses are often cross-protective. Thus the activation of Nrf2 should protect against a variety of biotic and abiotic stresses. To study the in vivo protective effects of avicins, we chose UVB light, an ancient stressor, a complete carcinogen, and the major cause of skin cancers. Avicins delivered topically, either before or immediately after exposure to UVB in SKH-1 mice, significantly suppressed skin damage. With regards to initiation, avicins were unable to diminish the UVB-induced base dimerization (data not shown), indicating the absence of a sunscreen effect. On the other hand, avicins induced a dramatic decrease in the generation of ROS as measured by levels of 8-OH-dG, which is a chemical signature of hydroxyl radical attack on DNA. Expression of phase 2 proteins like NQO1 and HO-1 was also enhanced in vivo. Since generation of ROS has been directly associated with DNA damage such as p53 mutagenesis (42), the significant suppression of p53 mutations probably occurs in part as a result of the antioxidant effects of avicins. The proapoptotic property of avicins could also account for the observed decrease in p53 mutations. The ability to remove damaged cells by apoptosis (1), an additional important form of stress response in multicellular organisms, suggests novel intervention strategies for premalignant and malignant skin and other diseases.

One of the important pathophysiological consequences of chronic UV exposure is depletion of antioxidants such as GSH, which is known to sensitize the skin to damage (43). Thus the ability of avicins to increase GST and TRX_red, which recycles ascorbic acid as well as other reducing substances (44), further supports the UV-protectant effects of avicins. Decrease in lipid peroxidation and increase in the levels of lipid-soluble antioxidants like bilirubin and in scavengers of lipid peroxides such as GSH and GST also protect the cells from UV-induced membrane damage.

The five-ring triterpene structures called hopanoids are ancient and have been identified as the main membrane-lipid support in a number of prokaryotes that evolved before the rise in atmospheric oxygen, when synthesis of sterols became possible (45). Hopanoids have a fascinating property of preventing the inward leakage of protons (46), which could contribute to a decrease in oxidative stress. The evolution of the side chain containing the two reactive electrophile sites converted the triterpene to a hybrid signaling molecule and probably first occurred in plants. Like ROS and reactive nitrogen species, reactive electrophilic metabolites such as avicins evolved as cell-signaling molecules with submolecular (atomic) specificity (47). We speculate that the avicins may regulate redox balance in plants too, based on the avicins’ identification in critical organs necessary for overall fitness, namely the seedpods, germinating seeds, and roots (2, 48). Like certain phenolic compounds known to prevent cellular damage (49), the avicins are induced when Acacia victoriae is under stress (48). The variety of defense molecules induced by avicins play critical roles in plant defense against oxidant stress (50).

In conclusion, avicins represent a new class of metabolites that regulate cellular redox balance through DNA-binding sensor proteins and that appear capable of modulating stress-associated disease phenotypes that are thermodynamically imbalanced (51). Based on the properties of avicins reported here and previously, we speculate that they mimic many of the effects of calorie restriction (49). The increased understanding of avicins’ effects on inflammatory tissue, including UV-exposed skin, offers exciting opportunities for disease modulation (52, 53, 54). It is hoped that the avicins, having moved from the Australian desert to the laboratory, will successfully find application at the bedside.

View Supplemental data

Research was supported by the Clayton Foundation for Research. Additional support was provided by the Biomedical Research Foundation and the Abraham J. and Phyllis Katz Foundation. We would like to thank Anil Jaiswal (Baylor College of Medicine) for providing some of the reagents.

Conflict of interest: The authors have declared that no conflict of interest exists.

Nonstandard abbreviations used: NF-E2–related factor 2 (Nrf2); antioxidant response element (ARE); avicin extracts (AE); heme oxygenase (HO); NADPH:quinone oxidoreductase 1 (NQO1); enhanced green fluorescence protein–Nrf2 plasmid (pEGFP-Nrf2); human ARE (hARE); glutathione-S-transferase (GST); glutathione peroxidase (GPx); glutathione (GSH); thioredoxin (TRX); TRX reductase (TRX_red); tert-butyl hydroquinone (tBHQ); 8-hydroxy-2′-deoxyguanosine (8-OH-dG); reactive oxygen species (ROS).

1. Haridas, V, et al. Avicins: triterpenoid saponins from Acacia victoriae (Bentham) induce apoptosis by mitochondrial perturbation. Proc. Natl. Acad. Sci. U. S. A. 2001. 98:5821-5826.
   View this article via: PubMed CrossRef Google Scholar
2. Mujoo, K, et al. Triterpenoid saponins from Acacia victoriae (Bentham) decrease tumor cell proliferation and induce apoptosis. Cancer Res. 2001. 61:5486-5490.
   View this article via: PubMed Google Scholar
3. Hanausek, M, et al. Avicins, a family of triterpenoid saponins from Acacia victoriae (Bentham), suppress H-ras mutations and aneuploidy in a murine skin carcinogenesis model. Proc. Natl. Acad. Sci. U. S. A. 2001. 98:11551-11556.
   View this article via: PubMed CrossRef Google Scholar
4. Haridas, V, Arntzen, CJ, Gutterman, JU. Avicins, a family of triterpenoid saponins from Acacia victoriae (Bentham), inhibit activation of nuclear factor-kappa B by inhibiting both its nuclear localization and ability to bind DNA. Proc. Natl. Acad. Sci. U. S. A. 2001. 98:11557-11562.
   View this article via: PubMed CrossRef Google Scholar
5. Talalay, P, De Long, MJ, Prochaska, HJ. Identification of a common chemical signal regulating the induction of enzymes that protect against chemical carcinogenesis. Proc. Natl. Acad. Sci. U. S. A. 1988. 85:8261-8265.
   View this article via: PubMed CrossRef Google Scholar
6. Dhakshinamoorthy, S, Jaiswal, AK. Small Maf (MafG and MafK) proteins negatively regulate antioxidant response element-mediated expression and antioxidant induction of the NAD (P) H:quinone oxidoreductase1 gene. J. Biol. Chem. 2000. 275:40134-40141.
   View this article via: PubMed CrossRef Google Scholar
7. Shaw, PM, et al. The human dioxin-inducible NAD(P)H: quinone oxidoreductase cDNA-encoded protein expressed in COS-1 cells is identical to diaphorase 4. Eur. J. Biochem. 1991. 195:171-176.
   View this article via: PubMed CrossRef Google Scholar
8. Holmgren, A, Bjornstedt, M. Thioredoxin and thioredoxin reductase. Methods Enzymol. 1995. 252:199-208.
   View this article via: PubMed CrossRef Google Scholar
9. Perez, P, et al. Altered skin development and impaired proliferative and inflammatory responses in transgenic mice overexpressing the glucocorticoid receptor. FASEB J. 2001. 15:2030-2032.
   View this article via: PubMed Google Scholar
10. Singh, RP, Tyagi, AK, Zhao, J, Agarwal, R. Silymarin inhibits growth and causes regression of established skin tumors in SENCAR mice via modulation of mitogen-activated protein kinases and induction of apoptosis. Carcinogenesis. 2002. 23:499-510.
    View this article via: PubMed CrossRef Google Scholar
11. Floyd, RA, Lewis, CA. Hydroxyl free radical formation from hydrogen peroxide by ferrous iron-nucleotide complexes. Biochemistry. 1983. 22:2645-2649.
    View this article via: PubMed CrossRef Google Scholar
12. Snedecor, G.W., and Cochran, W.G. 1967. Statistical methods. Iowa State University Press. Ames, Iowa, USA. 325 pp.
13. Itoh, K, et al. Keap 1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain. Genes Dev. 1999. 13:76-86.
    View this article via: PubMed CrossRef Google Scholar
14. Jaiswal, AK. Regulation of genes encoding NAD(P)H:quinone oxidoreductases. Free Radic. Biol. Med. 2000. 29:254-262.
    View this article via: PubMed CrossRef Google Scholar
15. Chui, DHK, Tang, WZ, Orkin, SH. cDNA cloning of murine Nrf 2 gene, coding for a p45 NF-E2-related transcription factor. Biochem. Biophys. Res. Commun. 1995. 209:40-46.
    View this article via: PubMed CrossRef Google Scholar
16. Dinkova-Kostova, AT, et al. Direct evidence that sulfhydryl groups of Keap1 are the sensors regulating induction of phase 2 enzymes that protect against carcinogens and oxidants. Proc. Natl. Acad. Sci. U. S. A. 2002. 99:11908-11913.
    View this article via: PubMed CrossRef Google Scholar
17. Poss, KD, Tonegawa, S. Reduced stress defense in heme oxygenase-1 deficient cells. Proc. Natl. Acad. Sci. U. S. A. 1997. 94:10925-10930.
    View this article via: PubMed CrossRef Google Scholar
18. Sabaawy, HE, et al. Human heme oxygenase-1 gene transfer lowers blood pressure and promotes growth in spontaneously hypertensive rats. Hypertension. 2001. 38:210-215.
    View this article via: PubMed Google Scholar
19. Ryter, SW, Tyrrell, RM. The heme synthesis and degradation pathways: role in oxidant sensitivity. Free Radic. Biol. Med. 2000. 28:289-309.
    View this article via: PubMed CrossRef Google Scholar
20. Stocker, R, Yamammoto, Y, McDonagh, AF, Glazer, AN, Ames, BN. Bilirubin is an antioxidant of possible physiological importance. Science. 1987. 235:1043-1046.
    View this article via: PubMed CrossRef Google Scholar
21. Torti, FM, Torti, SV. Regulation of ferritin genes and protein. Blood. 2002. 99:3505-3516.
    View this article via: PubMed CrossRef Google Scholar
22. Wasserman, WW, Fahl, WE. Functional antioxidant responsive elements. Proc. Natl. Acad. Sci. U. S. A. 1997. 94:5361-5366.
    View this article via: PubMed CrossRef Google Scholar
23. Halliwell, B. Oxidative stress in cell culture: an under-appreciated problem? FEBS Lett. 2003. 540:3-6.
    View this article via: PubMed CrossRef Google Scholar
24. McClintock, B. The significance of responses of the genome to challenge. Science. 1984. 226:792-801.
    View this article via: PubMed CrossRef Google Scholar
25. DePinho, RA, Wong, K-K. The age of cancer: telomeres, checkpoints, and longevity. J. Clin. Invest. 2003. 111(Suppl.):S9-S14.
    View this article via: JCI PubMed Google Scholar
26. Ramalho-Santos, M, Yoon, S, Matsuzaki, Y, Mulligan, RC, Melton, DA. “Stemness”: transcriptional profiling of embryonic and adult stem cells. Science. 2002. 298:597-600.
    View this article via: PubMed CrossRef Google Scholar
27. Brash, DE, Havre, PA. New careers for antioxidants. Proc. Natl. Acad. Sci. U. S. A. 2002. 99:13969-13971.
    View this article via: PubMed CrossRef Google Scholar
28. Soberman, RJ. The expanding network of redox signaling: new observations, complexities, and perspectives. J. Clin. Invest. 2003. 111:571-574. doi:10.1172/JCI200318099.
    View this article via: JCI PubMed Google Scholar
29. Kim, SO, et al. Oxy R: a molecular code for redox-related signaling. Cell. 2002. 109:383-396.
    View this article via: PubMed CrossRef Google Scholar
30. Itoh, K, et al. Keap1 regulates both cytoplasmic-nuclear shuttling and degradation of Nrf2 in response to electrophiles. Genes Cells. 2003. 8:379-391.
    View this article via: PubMed CrossRef Google Scholar
31. Halliwell, B., and Gutteridge, J.M.C. 1998. Free radicals, other reactive species and disease. In Free radicals in biology and medicine. B. Halliwell and J.M.C. Gutteridge, editors. Oxford University Press. Oxford, United Kingdom. 617–783.
32. Smith, MT. Benzene, NQO1, and genetic susceptibility to cancer. Proc. Natl. Acad. Sci. U. S. A. 1999. 96:7624-7626.
    View this article via: PubMed CrossRef Google Scholar
33. Hegyi, T, Goldie, E, Hiatt, M. The protective role of bilirubin in oxygen-radical diseases of the preterm infant. J. Perinatol. 1994. 14:296-300.
    View this article via: PubMed Google Scholar
34. Yachie, A, et al. Oxidative stress causes enhanced endothelial cell injury in human heme-oxygenase-1 deficiency. J. Clin. Invest. 1999. 103:129-135.
    View this article via: JCI PubMed CrossRef Google Scholar
35. Chan, K, Han, X-D, Kan, YW. An important function of Nrf2 in combating oxidative stress: detoxification of acetaminophen. Proc. Natl. Acad. Sci. U. S. A. 2001. 98:4611-4616.
    View this article via: PubMed CrossRef Google Scholar
36. Lei, XG. In vivo antioxidant role of glutathione peroxidase: evidence from knockout mice. Methods Enzymol. 2002. 347:213-225.
    View this article via: PubMed CrossRef Google Scholar
37. Brooks, JD, Goldberg, MF, Nelson, LA, Wu, D, Nelson, WG. Identification of potential prostate cancer preventive agents through induction of quinone reductase in vitro. Cancer Epidemiol. Biomarkers Prev. 2002. 11:868-875.
    View this article via: PubMed Google Scholar
38. Shih, AY, et al. Coordinate regulation of glutathione biosynthesis and release by Nrf2-expressing glia potently protects neurons from oxidative stress. J. Neurosci. 2003. 23:3394-3406.
    View this article via: PubMed Google Scholar
39. Motterlini, R, et al. Endothelial heme-oxygenase-1 induction by hypoxia: modulation by inducible nitric oxide synthase and S-nitrosothiols. J. Biol. Chem. 2000. 275:13613-13620.
    View this article via: PubMed CrossRef Google Scholar
40. Almeras, E, et al. Reactive electrophile species activate defense gene expression in Arabidopsis. Plant J. 2003. 34:205-216.
    View this article via: PubMed CrossRef Google Scholar
41. Farmer, EE. Surface-to-air signals. Nature. 2001. 411:854-856.
    View this article via: PubMed CrossRef Google Scholar
42. Yamanishi, Y, et al. Regional analysis of p53 mutations in rheumatoid arthritis synovium. Proc. Natl. Acad. Sci. U. S. A. 2002. 99:10025-10030.
    View this article via: PubMed CrossRef Google Scholar
43. Tyrrell, RM, Pidoux, M. Correlation between endogenous glutathione content and sensitivity of cultured human skin cells to radiation at defined wavelengths in the solar ultraviolet range. Photochem. Photobiol. 1988. 47:405-412.
    View this article via: PubMed CrossRef Google Scholar
44. May, JM, Cobb, CE, Mendiratta, S, Hill, KE, Burk, R. Reduction of the ascorbyl free radical to ascorbate by thioredoxin reductase. J. Biol. Chem. 1998. 36:23039-23045.
45. Sahm, H, Rohmer, M, Bringer-Meyer, S, Sprenger, GA, Welle, R. Biochemistry and physiology of hopanoids in bacteria. Adv. Microb. Physiol. 1993. 35:247-273.
    View this article via: PubMed CrossRef Google Scholar
46. Haines, TH. Do sterols reduce proton and sodium leaks through lipid bilayers? Prog. Lipid Res. 2001. 40:299-324.
    View this article via: PubMed CrossRef Google Scholar
47. Nathan, C. Specificity of a third kind: reactive oxygen and nitrogen intermediates in cell signaling. J. Clin. Invest. 2003. 111:769-778. doi:10.1172/JCI200318174.
    View this article via: JCI PubMed Google Scholar
48. Joshi, L, et al. Metabolomics of plant saponins: bioprospecting triterpene glycoside diversity with respect to mammalian cell targets. OMICS. 2002. 6:235-256.
    View this article via: PubMed CrossRef Google Scholar
49. Howitz, KT, et al. Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan. Nature. 2003. 425:191-196.
    View this article via: PubMed CrossRef Google Scholar
50. Vranova, E., Van Breusegem, F., Dat, J., Belles-Boix, E., and Inze, D. 2002. The role of active oxygen species in plant signal transduction. In Plant signal transduction. D. Scheel and C. Wasternack, editors. Oxford University Press. Oxford, United Kingdom. 45–73.
51. Strohman, R. Thermodynamics: old laws in medicine and complex disease. Nat. Biotechnol. 2003. 21:477-479.
    View this article via: PubMed CrossRef Google Scholar
52. Coussens, LM, Werb, Z. Inflammation and cancer. Nature. 2002. 420:860-867.
    View this article via: PubMed CrossRef Google Scholar
53. Nathan, C. Points of control in inflammation. Nature. 2002. 420:846-852.
    View this article via: PubMed CrossRef Google Scholar
54. Lane, N. 2002. Life, death and oxygen: lessons from evolution on the future of ageing. In Oxygen: The Molecule That Made the World. N. Lane, editor. Oxford University Press. Oxford, United Kingdom. 314–342.