Activation of genes involved in xenobiotic metabolism is a shared signature of mouse models with extended lifespan

Abstract

Xenobiotic metabolism has been proposed to play a role in modulating the rate of aging. Xenobiotic metabolizing enzymes (XME) are expressed at higher levels in calorically restricted mice (CR) and in GH/IGF-I-deficient, long-lived mutant mice. In this study, we show that many phase I XME genes are similarly upregulated in additional long-lived mouse models, including "crowded litter" (CL) mice, whose lifespan has been increased by food restriction limited to the first 3 wk of life, and in mice treated with rapamycin. Induction in the CL mice lasts at least through 22 mo of age, but induction by rapamycin is transient for many of the mRNAs. Cytochrome P-450s, flavin monooxygenases, hydroxyacid oxidase, and metallothioneins were found to be significantly elevated in similar proportions in each of the models of delayed aging tested, whether these were based on mutation, diet, drug treatment, or transient early intervention. The same pattern of mRNA elevation could be induced by 2 wk of treatment with tert-butylhydroquinone, an oxidative toxin known to activate Nrf2-dependent target genes. These results suggest that elevation of phase I XMEs is a hallmark of long-lived mice and may facilitate screens for agents worth testing in intervention-based lifespan studies.

Fig. 1.

Scatterplots of mRNA levels for enzymes involved in phase I detoxification of xenobiotics in liver, expressed as ratios to their respective control mice, for crowded litter (CL) mice compared with calorie-restricted (CR) mice (A), Snell mice (B), rapamycin-treated mice (C), and published effects of exposure to CA (D) (1). Each symbol represents a different mRNA. All mice were 12 mo old, except 2–4 mo for the CA group. Each panel also shows the Spearman correlation coefficient R and its associated P value. Calculated values are means of n = 6 experimental mice compared with n = 6 of the relevant age-matched control. Diagonal lines show best least squares linear fit. Note that vertical axis scales differ among panels to provide legible detail for each of the data sets compared.

Fig. 2.

CL mice maintain elevated levels of liver phase I mRNAs through 22 mo of age. Each panel shows 1 of the 6 mRNA species that was significantly elevated in 12-mo-old CL mice. Bars indicate means ± SE for CL and age-matched control mice, in each case normalized to the average level of mRNA in 12-mo-old control animals; n = 6 mice per treatment, except n = 5 for 22-mo-old CL mice. Asterisks over CL bars indicate significant differences from age-matched controls (2-tailed t-test, *P < 0.05, **P < 0.01). Horizontal lines indicate significant differences vs. levels in 12-mo-old control mice.

Fig. 3.

Rapamycin-dependent elevation of many xenobiotic genes is reversed later in life. Mice were given food containing rapamycin starting at 9 mo of age and euthanized at 12 or 22 mo of age. Bars indicate means ± SE for rapamycin-treated and age-matched control mice, in each case normalized to the average level of mRNA in 12-mo-old control animals; n = 6 mice per treatment. Asterisks over rapamycin bars indicate significant differences from age-matched controls (2-tailed t-test, *P < 0.05, **P < 0.01). Horizontal lines indicate significant differences vs. 12-mo-old control mice or effects of age, as indicated.

Fig. 4.

Acute exposure to tert-butylhydroquinone (tBHQ) mimics mRNA patterns seen in CR, CL, Snell, and cholic acid (CA)-treated mice but not rapamycin-treated mice. Scatterplots show mRNA levels for xenobiotic metabolism genes in liver of mice after 2-wk exposure to tBHQ, compared with CR (A), CL (B), Snell (C), rapamycin-treated (D), and CA-treated mice (E). Each symbol represents a different mRNA, plotted to show the ratio of the mean level in experimental mice to their respective controls. All mice were 12 mo old except for the published data on CA (1), which represent mice 2–4 mo of age. Each panel also shows the Spearman correlation coefficient R with its associated P value. Diagonal lines indicate least square fits. Note that the correlation of tBHQ effects with rapamycin is not statistically significant.

Fig. 5.

Correlation matrix of hepatic xenobiotic metabolism gene expression. Each box shows the Spearman correlation coefficient R for the populations of mice in the corresponding row and column header. Values of R > 0.5 are statistically significant at P < 0.05. Dark gray shading denotes an R value > 0.8, light gray denotes an R value ranging from 0.5 to 0.8. Note that no significant relationship was observed for comparisons of rapamycin and tBHQ, Snell, GHRKO, or Little mice. CA and Little data were obtained from a previously published study (1).

Fig. 6.

Model for Nrf2-mediated regulation of aging. Nrf2 is a stress-responsive transcription factor that regulates expression of multiple protective genes, including XMEs. Activation of XMEs may protect cells from damage and could, in principle, help delay aging. Nrf2 is negatively regulated by Keap1 protein, which promotes rapid turnover of Nrf2 by the proteasome. Keap1 contains multiple cysteine residues, which are oxidized under stressful conditions, or by chemical Nrf2 activators, such as tBHQ, resulting in conformational changes that release bound Nrf2. Nrf2 can then translocate to the nucleus and induce transcription of XMEs through antioxidant response elements (ARE) and may also interact with nuclear receptors (NR) or xenobiotic response elements (XRE). Nrf2 may also be repressed by GH/IGF-I signaling or the target of rapamycin (TOR) pathway. CR and rapamycin treatment may enhance Nrf2 activity through attenuation of these signaling pathways.

Fig. 7.

Calculated values are means of gene expression as a ratio of experimental mice to age-matched control mice, normalized to β-actin levels. CR, calorie restricted; CL, crowded litter; Rapa, rapamycin treated; tBHQ, tert-butylhydroquinone treated; CA, cholic acid treated; GHRKO, GH receptor knockout. Degrees of shading indicate increasing levels of gene expression. Boldface indicates P < 0.05.