Does Bach1 & C-Myc Dependent Redox Dysregulation of Nrf2 & Adaptive Homeostasis Decrease Cancer Risk in Ageing?

Abstract

The Keap1-Nrf2 signal transduction pathway plays a major role in oxidant and electrophile induction of adaptive homeostasis that transiently and reversibly increases cellular and organismal protection from stress. By expanding (and then contracting) the normal homeostatic range of expression of stress-protective genes, Nrf2 allows us to cope with fluctuations in stress levels. Two major inhibitors of Nrf2 are Bach1 and c-Myc which normally serve the important function of turning off adaptation when appropriate. We have found, however, that both Bach1 and c-Myc levels increase substantially with age and that older human cells, worms, flies, and mice loose Nrf2-dependent signaling and adaptive homeostasis. Nrf2 has also been linked with increased risk of cancers, and cancer incidence certainly increases with age. Here we propose that the age-dependent increase in Bach1 and c-Myc may actually cause the age-dependent decline in Nrf2 signaling and adaptive homeostasis, and that this is a coordinated attempt to minimize the age-dependent increase in cancer incidence. In other words, we may trade off adaptive homeostasis for a lower risk of cancer by increasing Bach1 and c-Myc in ageing.

Graphical Abstract

Origins of the Free Radical Theory of Ageing

From today’s perspective, it can seem hard to imagine that free radicals, oxidative stress, and redox regulation have not always been commonly accepted elements of biological systems. In actuality, however, the very idea that free radical reactions could be widely experienced by living organisms took a long time to be accepted by mainstream scientists.

It is always difficult to determine exactly who made ‘the most important early discoveries’ that helped launch any given field, but with apologies for any omissions (or even excessive praise) we have attempted to list at least some of the groundbreaking early contributions to the free radical biology & medicine field. In 1894 Harry Fenton [1] discovered the basis for what has come to be known as the ‘Fenton Reaction’ when he showed that hydrogen peroxide could oxidize ferrous sulfate to generate a species that, in turn, would oxidize tartaric acid. Then, in 1900, Moses Gomberg [2] considered for the first time that triphenyl methyl radicals could play significant roles in living systems. More than 50 years later, in 1954, Rebecca Gershman [3] proposed that the damaging effects of X radiation and the phenomenon of ‘oxygen poisoning’ shared a common mechanism involving free radicals.

In the same year, Barry Commoner [4] provided direct evidence of free radicals in biological systems using electron paramagnetic resonance spectroscopy.

In 1956, just two years after Gershman and Commoners important papers, Denham Harman [5], working at the University of California at Berkeley, made a truly amazing leap in proposing that free radical damage to “ …..cell constituents and on the connective tissues.” could actually underlie the ageing phenomenon. It should be noted that when Harman proposed his ‘Free Radical Theory of Aging,’ uncatalyzed one-electron oxidation/reduction reactions were still not widely considered to be of biological importance. In fact, it was to take another 13 years until Joe McCord and Irwin Fridovich [6] could demonstrate that an enzyme encoded by a specific gene is utilized to begin the detoxification of the superoxide anion radical (O_2^•−), in discovering the function of superoxide dismutase. This seminal discovery opened a floodgate of investigations into free radical biology and oxidative stress that still continues to this day.

Free Radical Toxicity and Antioxidant Compounds

Ever since the 1950’s, a major focus of free radical biology has been the toxicity of radicals like O_2^•−, hydroxyl radicals (^•OH), peroxyl radicals (ROO^•), peroxynitrite (NOO^•); and related oxygen- and nitrogen-based oxidants such as hydrogen peroxide (H₂O₂), singlet oxygen ¹O₂, ozone (O3), and lipid hydroperoxides (ROOH). Such species have clearly been shown to be generated by various metabolic pathways in vivo, and are also common environmental toxicants. In addition, many medically useful drugs and diagnostic tools, such as X ray scans, involve significant exposure to reactive oxygen and nitrogen species. As a result, an enormous literature in free radical biology & medicine has focused largely on oxidative damage to cell structures, proteins, lipids, and DNA, and the effects such exposures may have on disease risk and lifespan.

Once a link between oxidation and toxicity, disease, and even death was considered feasible, researchers began to look for antioxidants that might ameliorate the problem. Numerous plant-based molecules that have clear antioxidant properties, at high concentrations in test tube reactions, have been proposed as ‘healthy’ dietary supplements over the years. The concept is that such molecules act as ‘suicide substrates’ or ‘sacrificial lambs,’ by being oxidized themselves (and then eliminated) to protect cellular structures, proteins, lipids, and DNA. With the exception of vitamin E (α tocopherol), however, which does appear to exert significant protection as a chain-breaking antioxidant in lipid membranes, no other dietary antioxidant supplement has been shown to exert significant direct antioxidant effects in vivo. The problem is quite simply one of concentration. Although reaction rates for biologically relevant reactive oxygen and nitrogen species vary widely, metabolites including amino acids, carbohydrates, and lipids, cell proteins, and DNA typically react with such species at the same or very similar rates as do dietary antioxidants. Thus, for a dietary supplement to be effective as a direct antioxidant, it would have to reach intracellular concentrations comparable to those of our metabolites, proteins, lipids, and DNA - something that is neither conceivable nor desirable [7]!

‘Antioxidant’ Enzymes and Damage Removal/Repair Systems

Of course, antioxidant enzymes such as the superoxide dismutases, glutathione peroxidases glutathione reductases, glutathione transferases, peroxiredoxins, and (in some situations) catalase, etc. can be effective at quite low concentrations because they catalyze detoxification reactions at rates that are several orders of magnitude greater than non-catalytic, simple antioxidant dietary molecules can achieve.

Important extra layers of protection from oxidation have become apparent over the years with the discovery of damage removal and/or repair systems for DNA, proteins, and even lipids [8]. Numerous DNA repair systems for both single strand oxidative damage and even double strand oxidative damage have been elucidated, and such systems clearly play major roles in maintaining the integrity and fidelity our genomes [9]. Enzymes such as methionine sulfoxide reductase have been shown to re-reduce oxidized protein sulfoxide residues to restore function [10], and proteinases such as the Proteasome, the Immunoproteasome, and the mitochondrial Lon protease exert major intracellular protective roles by proteolytic degradation of oxidized proteins which minimizes their aggregation and cross-linking [11][12]. Finally, phospholipase A2 and phospholipase C have been shown to selectively cleave (per)oxidized lipids from biological membranes, thus beginning the process of membrane repair [13]. Clearly, these damage removal and repair systems form a major part of the reason that we are able to exist for some 70 – 100 years in an oxidizing environment.

Not all Cells, Tissues, and Organs Age in the Same Way

When pondering the ageing phenomenon in multicellular organisms, it is always important to remember that divisionally competent cells and non-dividing cells have very different life histories. Thus, microglia and neurons experience ageing in very different ways. Similarly, the ageing process in rapidly dividing skin cells is radically different from ageing in non-dividing skeletal and cardiac muscle cells. Divisionally competent cells have been shown to protect themselves from oxidative stress by halting growth and division and supercoiling their DNA and coating it with protective proteins in a process initiated by growth arrest and DNA damage (GADD) genes, such as GADD45, GADD153, and GADD7 [14]. Non-dividing cells may be inherently better-protected against oxidants since their DNA spends relatively little time in an uncoiled and vulnerable state.

Oxidative Stress and Problems with the Free Radical Theory of Ageing

Much of what we have been discussing thus far would come under the heading of ‘oxidative stress,’ a term originally coined by Helmut Sies [15]. In a subsequent publication Sies [16], went on to state, “The imbalance between oxidants and antioxidants in favor of the oxidants, potentially leading to damage, forms the core of the definition of ‘oxidative stress’, as this author has discussed at various occasions before and after the 1985 book devoted to the term itself.” Thus, oxidative stress, as originally understood, implies that cells, tissues, organs, or organisms are under such pressure from oxidative insults that overwhelm the antioxidant defenses that they may cease to function properly, or even die. Oxidative stress has been such a major concept in the free radical field that it has dominated much of the literature.

In recent years, several important developments and discoveries have prompted new ways of thinking about free radicals, oxidative stress, and ageing, and a reevaluation of the interactions between them. Firstly, careful studies of oxidative damage over the course of a lifetime (as recently reviewed [17])have revealed significant disagreements with the Free Radical Theory of Ageing as originally proposed by Harman [5]. Perhaps most significantly, Harman [5] suggested that oxidative damage to cell structures and components increased gradually over a lifetime to the point where function would be compromised. In point of fact, it has been found that, in multiple diverse organisms, damage accumulation, as evidenced by oxidized and aggregated proteins, is only significant in the last third of lifespan where it increases exponentially [18]. Presumably, this means that either the rate or extent of damage increases significantly in the last third of life, or that antioxidant defenses and damage removal & repair systems decline significantly in effectiveness, or that all of these occur. Being able to pinpoint which of these possibilities actually pertains would certainly be an important step forward in understanding the basic biology of ageing and would also help inform possible mechanistic approaches to alleviating age-related damage and senescence.

Transient Adaptation as an Oxidant Defense

From the 1970’s through the 1990’s, various groups discovered previously unknown links between reactive oxygen species and physiological adaptations. Adaptation of rats to hyperoxia was shown to involve increases in superoxide dismutase that declined when the animals were brought back to normoxia [19]. Scientists from Bruce Ames’ [20,21] laboratory demonstrated that bacteria could transiently adapt to normally toxic levels of H₂O₂ by increasing expression of protective & repair genes. Similarly, Greenberg et al. [22] and Wu & Weiss [23] reported that upregulated expression of key genes was responsible for reversible adaptation to O_2^•−. Studying a rodent model of physiological adaptation to exercise [24,25], Davies, Quintanilha, Brooks, Hochstein, and Packer [26,27] proposed that free radicals generated during endurance-type exercise actually might form the signal for the massive mitochondrial biogenesis (involving increased expression of hundreds of genes) that characterizes endurance exercise training. Subsequently, Davies, Lowry, and Davies [28] demonstrated that increased expression of protective genes was responsible for transient adaptation to H₂O₂ in yeast, and Weise, Pacifici, and Davies [29] confirmed the same phenomenon in mammalian cells.

In bacteria the oxyR regulon was shown to be responsible for mediating increased expression of multiple antioxidant and repair genes in response to exposure to H₂O₂ [20,21], whereas the bacterial soxRS regulon was found to perform a similar function in response to O_2^•− or redox cycling agents [22,23]. A bacterial regulon involves a relatively simple signaling system in which a group of genes is regulated as a single unit by a ‘sensing’ protein that responds to agents, such as H₂O₂ or O_2^•−, by activating (or repressing) expression of all genes in the group. Regulons such as oxyR and soxRS do not exist in eukaryotic cells but more complex systems, involving multiple interacting proteins, have been found to mediate redox regulation of gene expression. These complex protein-protein and protein-DNA interacting systems are called signal transduction pathways.

Beginning in the 1990’s, several laboratories demonstrated that oxidants could act as physiological second messengers in cell signaling [30,31]. Later, Dean Jones [32] and others proposed that the actual redox state of the cell may directly regulate cellular functions and stress resistance. These, and many other discoveries, lead to the concept that reactive oxygen/nitrogen species and/or redox state could regulate gene expression, resistance to stress, and physiological adaptations: this is the concept of “Redox Regulation.’ These findings and evolving concepts lead Sies [33] to reconfigure his concept of oxidative stress with the addition of redox signaling and control as “An imbalance between oxidants and antioxidants in favor of the oxidants, leading to a disruption of redox signaling and control and/or molecular damage.”

Hormesis versus Adaptive Homeostasis

Central to the concept of redox regulation is the principle that cells, tissues, organs, and organisms can transiently adapt to varying redox states. Some investigators have proposed that a small amount of oxidative damage results in repair and activation of protective systems exceeding the immediate need and that this phenomenon, called ‘Hormesis,’ underlies adaptive phenomena. While there is evidence for adaptive DNA repair following measurable DNA damage [9] there is little to suggest that Hormesis could explain non-toxic, physiological adaptations by redox regulation. Indeed, significant transient and reversible adaptations in gene expression and stress protection have been shown to result from very minor changes in oxidant or free radical exposure that are insufficient to cause any damage. It is now clear that many adaptive responses are induced best at very low, non-damaging, levels of redox signals, and that such adaptations become progressively less effective if exposure to higher levels of oxidants that also cause progressive increases in damage is employed [34]. These findings led to the concept of ‘Adaptive Homeostasis’ [35] in which regulation of gene expression by signal transduction pathways is considered to be a normal component of physiological homeostasis, rather than a phenomenon of toxicology. Basically, Adaptive Homeostasis postulates that the physiological range of homeostasis for gene expression and stress protection is under constant change; that is to say that the range is continuously being expanded or contracted based on prevailing signals, such as redox status or oxidant exposure. Adaptive Homeostasis has been defined thus, “The transient expansion or contraction of the homeostatic range in response to exposure to sub-toxic, non-damaging, signaling molecules or events, or the removal or cessation of such molecules or events “ [35].

The Keap1-Nrf2 Signal Transduction Pathway in Adaptive Homeostasis

The system which perhaps best exemplifies redox regulation and Adaptive Homeostasis is the Keap1-Nrf2 signal transduction pathway. A companion article in this mini-review collection gives a detailed description of the Keap1-Nrf2 pathway [36]. Excellent reviews of the significance of the proteins encoded by vitagenes and other protective enzymes regulated by Nrf2 have been recently published recently [37][38], so we shall limit ourselves to just a brief outline here. Basically, Nrf2 (nuclear factor erythroid 2 (NFE2)-related factor 2) resides in the cytoplasm of eukaryotic cells where it is bound to a multi-protein system involving Keapl (Kelch-like erythroid cell-derived protein with CNC homology-associated protein 1) and a Cul3-Rbx1 E3 ubiquitin ligase which rapidly polyubiquitinylates Nrf2 as it binds to the complex. The polyubiquitin chain targets Nrf2 for degradation by the 26S Proteasome and, thus, ensures a low steady-state cytoplasmic Nrf2 concentration. In the presence of signaling agents including H₂O₂ or electrophiles, such as 4-hydroxynonenal, key cysteine residues in Keapl are oxidized or alkylated allowing Nrf2 to avoid degradation, undergo phosphorylation and transit into the cell nucleus. The 26S Proteasome also undergoes disassembly in the presence of H₂O₂ that prevents Nrf2 degradation and rapidly increases the pool of Nrf2 available for nuclear transfer [39].

Once in the nucleus, Nrf2 binds to consensus sequences called ARE (Antioxidant Response Elements) or EpRE (Electrophile Response Elements) in the upstream control regions of multiple defense, damage removal, and damage repair genes and increases their transcription [37] Over time, the levels of Nrf2 bound to ARE/EpRE sites declines as competitors such as Bach1 (BTB domain and CNC homolog 1) displace it, initiating degradation, which c-Myc facilitates, thereby returning levels of gene expression to pre-signaling levels. This process may be seen as a transient expansion of the homeostatic range of gene expression mediated by the Keap1-Nrf2 signal transduction pathway, followed by a contraction of gene expression levels mediated by Bach1/c-Myc: in other words, an example of Adaptive Homeostasis in action. What should also be clear from the above discussion is that no damage or excessive repair (i.e. no Hormesis) is involved in any way. Instead, the Nrf2 signaling molecule undergoes a discreet post-translational modification that allows it to perform its designated function in the nucleus.

Normal Physiological Role of Bach1

First, Bach1 should not be confused with BACH1 (BRCA1-associated C-terminal helicase 1)[40]. Bach1 that is of concern here, is best known as a repressor of heme oxygenase 1 expression, which is induced by heme [41]. Bach1 is also known as a repressor of the expression of many other Nrf2-regulated genes [42]. Bach1 appears to compete with Nrf2 for binding to the same EpRE cis-element in several genes [43] and Bach1 binding to DNA is decreased by electrophiles [44]. In contrast to the activation of Nrf2 in which electrophiles bind to Keap1, electrophiles bind directly to Bach1 [44]. But differences among Nrf2-regulated genes in the effect of Bach1 silencing, suggest differential binding of Bach1 and Nrf2 to EpRE sequences [45,46]. Indeed consensus EpRE sequences vary considerably in their roles in gene regulation; e.g., of the thirteen consensus EpREs in human GCLC, three are involved in basal expression and only one is responsive to electrophiles [47]. Thus, it is possible that, in the terms of kinetic competition, Bach1 may sometimes be either a non-competitive rather than competitive inhibitor or not an inhibitor of Nrf2 signaling at all. This clearly is an area that requires further exploration. Nonetheless, the increase in Bach1 in ageing is likely to contribute to age-related changes in expression and inducibility of many Nrf2-regulated genes.

Normal Physiological Role for c-Myc

The well-established role of c-Myc in normal physiology is as a transcription factor that binds to the consensus E-box (CACGTG) sequence along with its partner, Max, to activate transcription of a large number of genes including important cell proliferation regulating proteins [48,49]. Several years ago, a role for c-Myc in inhibiting Nrf2-regulated transcription was discovered in our laboratory [50]. Instead of competing with Nrf2 for the EpRE binding sites on DNA, c-Myc bound to Nrf2, which not only interfered with transcription, but appeared to facilitate the turnover of nuclear Nrf2 protein in the absence of electrophilic activation of Nrf2 [51]. Another study confirmed the direct interaction of c-Myc and Nrf2 bound to EpRE, but also showed that knockdown of c-Myc increased Nrf2 mRNA [52]. When electrophilic Nrf2 activators were added to human bronchial epithelial cells, nuclear c-Myc protein declined [51].

Declining Nrf2 Effectiveness in Adaptive Homeostasis During Ageing is Mirrored by Increasing Bach 1 and C-Myc, which are Known Inhibitors of Nrf2-Regulated Transcription.

In studies of the effect of chronic exposure to nanoparticles (nPM) present in ambient air, we showed that Nrf2 regulated gene expression increased in the lungs, liver and cerebellum of young male mice, but that middle-aged mice showed no inducibility [53]. The four Nrf2-regulated genes that we examined were the catalytic and modulatory subunits of glutamate cysteine ligase, GCLC and GCLM respectively, heme oxygenase 1 (HO-1), and NAD(P)H:quinone oxidoreductase 1 (NQO1). Interestingly, the basal expression of all four Nrf2 regulated genes, was increased in the middle-aged male mice. The same pattern of changes in baseline and inducibility in GCLC and GCLM as well as subunits of the proteasome were observed in female mice exposed to the same protocol [54]. In primary human bronchial epithelial cells, we found a similar pattern of increased basal expression of Nrf2-regulated genes along with a decline in inducibility by sulforaphane in cells from older donors [55]. A similar loss of inducibility of Nrf2 in mammals with age [56–60] has been reported by others, but these reports showed more variability in the effect of age on basal expression of Nrf2-regulated genes.

Nrf2 homologues such as CncC [61] in Drosophila melanogaster and SKN-1 [62] in Caenorhabditis elegans, also show a decline in signaling by their Nrf2 homologues. From this analysis, it appears that the basal expression of Nrf2 varies in ageing among species and tissues, but that the inducibility of Nrf2-regulated genes generally declines in ageing. Nonetheless, the underlying mechanism for the decline in inducibility of Nrf2-regulated genes has been unclear. One potential explanation lies in the observation that in both the male and female mice and in human bronchial epithelial cells, we found that the levels of two inhibitors of Nrf2-regulated transcription, Bach1 and c-Myc increased with age [53][54][55]. Interestingly, in emphysema, an age-related disease, patients have increased Bach1 in their lungs and alveolar macrophages compared with controls [63].

Cancer Risk Increases With Age: Possible Relationship to the Balance of Nrf2 with Bach1 and C-Myc.

The earliest work on understanding of Nrf2 regulation came from its discovery as a transcription factor that was induced by agents proposed for cancer chemoprevention [64] and was the essential transcription factor that bound to the electrophilic response element (EpRE), also known by the misnomer, antioxidant response element (ARE) [65][66]. Confirmation came from knocking out Nrf2 in mice, which caused increased carcinogenesis and reduced the efficient induction of anti-carcingogenic proteins [67]. Nrf2 inducible anti-carcinogenic proteins include NQO1, GCLC, GCLM, and some of the glutathione S-transferases. Thus, it was disconcerting when the observation was made that Nrf2 might promote some cancers [68] and that high baseline levels of Nrf2 were found in several cancers [69]. As Nrf2 is clearly critical for resistance of cells to oxidative and xenobiotic toxicity, increased Nrf2 would help cancer cells resist natural and chemotherapeutic attacks; however, as Nrf2 is also involved in other aspects of metabolism, the cancer promoting action of Nrf2 may not simply involve a chemoprotective role [69].

It is well established that the incidence of cancer increases with age [70]. It has occurred to us that the increases in Bach1 and c-Myc in ageing might serve the purpose of suppressing cancer in those cells that express high constitutive Nrf2 levels. In other words, maybe there would be an even greater increase in cancer with age without the possible protective effect of Bach1 and c-Myc in suppressing Nrf2-dependent adaptive homeostasis, and preventing the tendency of Nrf2 to increase cancer incidence.

Nrf2/c-Myc Antagonism in Cancer

As with Nrf2, c-Myc may be a double-edged sword in cancer. While the increase in c-Myc with age may decrease Nrf2-regulated gene expression, there is constitutively high expression of c-Myc in some cancers. For example, in Burkitt lymphoma, c-Myc is overexpressed due to a chromosomal translocation that puts it under control of an active immunoglobulin promoter [71]. Amplification of c-Myc is also common in non-small cell lung cancers (NSCLC) [72]. Interestingly, dysfunctional Keap1 mutations are also common in NSCLC [73]. Thus, while there appears to be a relationship between Nrf2 regulation and c-Myc in cancer, whether this involves an attempt of cells to downregulate Nrf2 by increasing c-Myc or vice versa is unclear but deserves consideration. The multiple roles of c-Myc in normal physiology [48][49], complicate sorting out its potential role in countering the pro-cancer function of Nrf2. Nonetheless, we found that the c-Myc bound to Nrf2 was not phosphorylated [51] while the stabilization of c-Myc as a transcription factor in its normal physiological role requires its phosphorylation by ERK, which is often highly active in proliferating cells [74]. Thus, there may be some aspects other than expression of Nrf2 and c-Myc that may be further explored in relationship to cancer.

Nrf2/Bach1 Antagonism in Cancer

As pointed out earlier, Bach1 silencing does not affect all Nrf2-regulated genes uniformly [45]. Furthermore, it should be kept in mind that while its best known role is as a suppressor of Nrf2, Bach1 has additional Nrf2-unrelated roles in cell physiology [45][46]. Thus, we cannot be certain that Bach1 would suppress everything that Nrf2 is doing to promote cancer. Nonetheless, there is evidence that Bach1 can function as an anticancer protein. Bach1 antagonizes Nrf2 transcription of transketolase, an enzyme that plays a key role in redox homeostasis in cancer cells and proliferation and thereby acts as a cancer suppressor [75]. Bach1 suppression of HO-1 can also have an anticancer effect as demonstrated in human leukemic cells [76] while colorectal cancer had higher than normal Bach1 levels [77].

The picture of the relationships of Nrf2, c-Myc, and Bach1 to cancer is rather fuzzy at this time. This partially results from looking at cancer as one disease, which it clearly is not. The image only becomes clearer when looking at the effects of these proteins on specific genes in specific cancers. Thus, it is important to add these caveats when making generalities in dealing with this very important yet complex aspect of cancer and ageing. With that in mind, we conclude this section with the following: The increase in c-Myc and Bach1 in ageing is expected to counteract the effects of some Nrf2-dependent genes in cancer and that where Nrf2 increases cancer potential, c-Myc and Bach1 may play an anticancer role.

We have described here the possibility of Bach1 and c-Myc in regulating Nrf2. What also should be considered in advancing the factors limiting Nrf2 in aging are other competitors of Nrf2, including Nrf1, Nrf3, and Bach2, and the partners of Nrf2 in binding to DNA, small Mafs, c-Jun , and JunD that influence Nrf2-dependent transcription. While small Mafs appear to be the usual partner for many Nrf2 regulated genes [64,78,79], there is evidence suggesting that c-Maf and MafG may also act as inhibitors [80] and that members of the Jun family [81,82] and ATF4 [83] may also partner with Nrf2.

Conclusions

Here we propose that the age-dependent increase in Bach1 and c-Myc may actually cause the age-dependent decline in Nrf2 signaling and adaptive homeostasis, and that this is a coordinated attempt to minimize the age-dependent increase in cancer incidence. In other words we may trade-off adaptive homeostasis for a lower risk of cancer by increasing Bach1 and c-Myc levels in ageing.

Figure 1 –

Increased Nuclear Bach1 Inhibition of Nrf2 in Ageing:

Sun et al. [84] showed that Bach1 dissociates from DNA when Nrf2 heterodimerizes with small Maf, allowing Nrf2 to bind to the EpRE. The figure shows the competition for binding to EpRE is dependent on the relative abundance of Nrf2 and Bach1. In unstimulated young cells, Nrf2 regulated genes have low expression due to a high Bach1/Nrf2 ratio. In young cells, when Nrf2 is activated by electrophilic stimuli including H₂O₂, Nrf2 levels increase in the nucleus and overwhelms Bach1 inhibition. In unstimulated old cells, Nrf2 is already relatively high in the nucleus compared with young cells even though Bach1 is also elevated. In old cells, a further increase in electrophile stimulated Nrf2-dependent transcription is dampened compared to young cells. The age-induced increase in Bach1, which in some cells is further elevated by electrophiles, is probably responsible (at least in part) for what appears to be a ceiling effect for Nrf2 signaling in ageing.

Figure 2 -.

c-Myc Inhibition of Nrf2:

c-Myc binds to Nrf2 that is already bound to an EpRE sequence with its partner, a small Maf protein or member of the Jun family. This process inhibits transcription of target genes. c-Myc also facilitates the degradation of nuclear Nrf2.

Highlights.

• Electrophile-enhanced Nrf2 signaling efficiency and efficacy decreases with ageing.

• Nrf2 inhibitors, Bach1 and c-Myc appear to explain the declining Nrf2 role with age.

• Abrogated Nrf2 signaling reduces adaptive homeostasis in cancerous cells.

• Inability to adapt to stress makes cancerous cells better targets for immune suppression

• Thus, abrogated Nrf2 signaling with age may be a concerted effort to reduce cancer incidence.

Abbreviations:

O_2^•−

superoxide anion radical

^•OH

hydroxyl radical

ROO^•

peroxyl radicals

NOO^•

peroxynitrite

H₂O₂

hydrogen peroxide

¹O₂

singlet oxygen

O₃

ozone

ROOH

lipid hydroperoxides

Nrf2

nuclear factor erythroid 2 (NFE2)-related factor

Keap1

Kelch-like erythroid cell-derived protein with CNC homology-associated protein 1

Bach1

BTB domain and CNC homolog 1c