SEARCH

SEARCH BY CITATION

Keywords:

  • anti-aging;
  • antioxidant;
  • ARF;
  • cellular senescence;
  • DNA damage;
  • Ink4a;
  • life-span studies;
  • cancer;
  • aging;
  • tumor suppression;
  • p16Ink4a;
  • p53

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

The proteins encoded by the Ink4/Arf locus, p16Ink4a, p19Arf and p15Ink4b are major tumour suppressors that oppose aberrant mitogenic signals. The expression levels of the locus are progressively increased during aging and genome-wide association studies have linked the locus to a number of aging-associated diseases and frailty in humans. However, direct measurement of the global impact of the Ink4/Arf locus on organismal aging and longevity was lacking. In this work, we have examined the fertility, cancer susceptibility, aging and longevity of mice genetically modified to carry one (Ink4/Arf-tg) or two (Ink4/Arf-tg/tg) intact additional copies of the locus. First, increased gene dosage of Ink4/Arf impairs the production of male germ cells, and in the case of Ink4/Arf-tg/tg mice results in a Sertoli cell-only-like syndrome and a complete absence of sperm. Regarding cancer, there is a lower incidence of aging-associated cancer proportional to the Ink4/Arf gene dosage. Interestingly, increased Ink4/Arf gene dosage resulted in lower scores in aging markers and in extended median longevity. The increased survival was also observed in cancer-free mice indicating that cancer protection and delayed aging are separable activities of the Ink4/Arf locus. In contrast to these results, mice carrying one or two additional copies of the p53 gene (p53-tg and p53-tg/tg) had a normal longevity despite their increased cancer protection. We conclude that the Ink4/Arf locus has a global anti-aging effect, probably by favouring quiescence and preventing unnecessary proliferation.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

The Ink4a/Arf/Ink4b locus, hereby abbreviated as Ink4/Arf, encodes two members of the Ink4 family of cyclin-dependent kinase inhibitors, p16Ink4a and p15Ink4b, and a completely unrelated protein called p19Arf (Gil & Peters, 2006; Kim & Sharpless, 2006). The locus is among the most frequently inactivated loci in human cancer, which reflects its central role in preventing cancer development. The three products of the locus participate in key cellular anti-proliferative responses, namely, senescence and apoptosis. Proteins p16Ink4a and p15Ink4b inhibit the activity of the Cdk4,6/cycD kinases promoting Rb-mediated proliferative arrest; meanwhile, p19Arf inhibits the ubiquitin ligase Mdm2, thus promoting p53 stabilization (Gil & Peters, 2006; Kim & Sharpless, 2006). Current evidence indicates that mitogenic over-stimulation of primary normal cells in vitro results in upregulation of the locus and thereby decreased mitogenic response and, eventually, upon maximal activation of the locus, may result in apoptosis or senescence (Collado et al., 2007). At the organismal level, the Ink4/Arf locus is silent during development and post-natal life, becoming progressively expressed from adulthood until very old ages (Zindy et al., 1997; Krishnamurthy et al., 2004; Ressler et al., 2006).

Importantly, numerous independent genome-wide association studies using large human cohorts have linked polymorphisms in close vicinity to the Ink4/Arf locus with aging-associated frailty and with a variety of aging-associated diseases, such as myocardial infarction, type 2 diabetes and stroke [reviewed in Sharpless & DePinho (2007), Melzer (2008); see also Matarin et al. (2008)]. However, the functional significance of these polymorphisms remains to be elucidated. A number of recent studies have implicated p16Ink4a as an important negative regulator of tissue regeneration upon acute damage in the haematopoietic system, endocrine pancreas, skeletal muscle and fat (Janzen et al., 2006; Krishnamurthy et al., 2006; Molofsky et al., 2006; Baker et al., 2008). It should be borne in mind that these experimental systems probably involve acute mitogenic stimulation and high levels of induction of p16Ink4a, and this may not be predictive of the long-term impact of moderately increased p16Ink4a on physiological regeneration. On the other hand, studies from our laboratory and others, have implicated an anti-aging function of p19Arf through its ability to reinforce the activity of p53 thus favouring the elimination of damaged cells (Matheu et al., 2007; Baker et al., 2008; Matheu et al., 2008). In this context, we consider of high relevance to determine in a direct manner and under normal physiological conditions the global impact of the Ink4/Arf locus on aging and longevity.

Previous work from our laboratory has shown that a transgenic allele of the entire locus, Ink4a/Arf/Ink4b, behaves similarly to the endogenous allele (Matheu et al., 2004). In particular, the transgenic Ink4/Arf allele is minimally expressed in primary embryonic cells but responds to mitogenic over-stimulation and protects against neoplastic transformation by oncogenes. At the organismal level, the transgenic allele results in a modest increase (approximately 50%) in the expression levels of the three genes, Ink4a, Arf and Ink4b, compared to the wild-type situation, yet this translates into a significantly higher resistance to chemically induced cancer, as well as, to aging-associated spontaneous cancer (Matheu et al., 2004; Matheu et al., 2007). Therefore, we regard this transgene as a suitable tool to evaluate the effects of a modest and normally regulated increase in the function of the locus. In the present work, we report the fertility, cancer susceptibility, aging and longevity of mice carrying one or two transgenic copies of the entire Ink4a/Arf/Ink4b locus.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Male sterility in Ink4/Arf-tg/tg mice

Intercrosses between Ink4/Arf-tg mice (Matheu et al., 2004) were performed to generate the three genotypes of relevance for this study, namely, Ink4/Arf-wt, Ink4/Arf-tg and Ink4/Arf-tg/tg, all of them in a pure (> 99%) C57BL6 genetic background. Therefore, the mice under study here can be considered to carry, respectively, 2 (wt), 3 (tg) or 4 (tg/tg) gene doses of Ink4a, Arf and Ink4b. We have previously shown that Ink4/Arf-tg mice have moderately increased levels of expression of Ink4a, Arf and Ink4b in a number of tissues at old ages (Matheu et al., 2004). In turn, as expected, doubly transgenic old Ink4/Arf-tg/tg mice had even higher levels of expression of Ink4a and Arf in spleen and liver compared to singly transgenic Ink4/Arf-tg mice (Supplementary Fig. S1). The first noticeable phenotype of the Ink4/Arf-tg/tg mice was that the males were infertile, whereas the females had no obvious fertility defects (Fig. 1A). We initially considered the possibility that the insertion of the transgene into the genome could have inactivated a gene specifically involved in male fertility. To examine this, we generated mice homozygous for the transgenic allele but lacking the endogenous alleles, i.e. Ink4/Arf-KO;tg/tg. Interestingly, Ink4/Arf-KO;tg/tg mice of both sexes were fertile and yielded the same number of pups per litter as wt mice (Fig. 1A). These results exclude the possibility of an insertional mutagenic event responsible for the male infertility and suggest a direct effect of the Ink4/Arf dosage on male fertility. In support of the latter scenario, we observed a reverse correlation between Ink4/Arf gene dosage and, both, testis size and sperm count. In the case of Ink4/Arf-tg/tg mice, their average testis weight was 18% compared to wt mice (100%), and were totally devoided of spermatozoa (Fig. 1B and C; see also below Fig. 2). Ink4/Arf-tg mice had intermediate deficits in testis weight and sperm count compared to doubly transgenic mice (Fig. 1B and C). These results point to a direct impact of the Ink4/Arf gene on male fertility.

image

Figure 1. Male infertility in Ink4a/Arf-tg/tg mice. (A) The average litter size for crosses (n ≥ 5) between mice of the indicated sex and genotype. In the case of male Ink4/Arf-tg/tg mice, no pregnancies were obtained from a total of nine males. (B) Representative image of testis from mice (8-weeks-old) of the indicated genotypes (left). Weight of testes (n = 3 per genotype) (right). (C) Sperm count per epididymis from mice (8-week-old) of the indicated genotypes (n = 3 per genotype). Data are mean values ± SEM. Student's t-test relative to wt: *P < 0.05; **P < 0.01; ***P < 0.001.

Download figure to PowerPoint

image

Figure 2. Depletion of spermatogonia in Ink4/Arf-tg/tg mice. Representative images of testis from mice (4- to 8-weeks-old, n = 5 per genotype) of the indicated genotypes analysed by immunohistochemistry to detect the indicated proteins. Note that the p16 staining gives a non-specific signal in the interstitial Leydig cells. In the tubules, germ cells are positive for p16Ink4a and p19Arf, whereas Sertoli cells are negative. Tubules from Ink4/Arf-tg/tg mice lack germ cells and therefore are negative for p16Ink4a and p19Arf; these tubules only contain Sertoli cells as indicated by the Sertoli cell marker Sox9.

Download figure to PowerPoint

Histological analyses of testes from 2-month-old Ink4/Arf-tg/tg mice revealed severe abnormalities in the testes. In particular, atrophic seminiferous tubules lacking germ cells (Fig. 2). In normal testes, the proliferative marker Ki67 is characteristically present in spermatogonia, however, Ink4/Arf-tg/tg testes were completely negative for Ki67, further confirming the absence of germ cells in these mice (Fig. 2). In the case of singly transgenic Ink4/Arf-tg mice, Ki67-positive cells were present albeit at lower levels than in wt testes (Fig. 2). Importantly, spermatogonia were clearly positive for p16Ink4a and p19Arf by immunohistochemistry both in wt and in Ink4/Arf-tg testes, while Sertoli cells were negative (Fig. 2). (Note that the p16 staining gives a non-specific signal in the interstitial Leydig cells). Consistent with the fact that Ink4/Arf-tg/tg testes are devoided of germ cells, these testes also lacked p16Ink4a- or p19Arf-positive cells (Fig. 2). Finally, we used Sox9, a well-established marker of Sertoli cells (Sekido & Lovell-Badge, 2008), to determine the impact of the Ink4/Arf transgene on this cell type. As shown in Fig. 2, all tubules in wt testes showed the characteristic pattern of Sertoli cells interspersed with spermatogonia; singly transgenic Ink4/Arf-tg testes had a mixture of normal tubules together with others exclusively composed by Sertoli cells; and, finally, more dramatically, the only cells present in Ink4/Arf-tg/tg testes were Sox9-positive thus resembling the so-called ‘Sertoli-cell-only syndrome’ (Fig. 2). It is important to mention that similar phenotypes have been also observed in mice with genetic alterations that upregulate the retinoblastoma (Rb) tumour suppressor pathway (see Discussion).

Increased cancer resistance in Ink4/Arf-tg/tg mice

We have reported that a single transgenic copy of the Ink4/Arf locus confers significant cancer protection to Ink4/Arf-tg mice compared to their wt counterparts (Matheu et al., 2004). We wondered whether cancer protection by the Ink4/Arf locus could be further increased in doubly transgenic Ink4/Arf-tg/tg mice. Cohorts of wt, tg and tg/tg mice were observed during their complete lifespan, sacrificed when moribund at old age, and subjected to detailed necropsy and histopathological analysis. Notably, almost half (48%) of the wt mice presented spontaneous malignant tumours, and this incidence was decreased to 38% in Ink4/Arf-tg mice (Fig. 3A). Remarkably, the presence of two transgenic copies resulted in potent cancer protection decreasing the incidence of aging-associated cancer to 10% in Ink4/Arf-tg/tg mice (Fig. 3A). These results demonstrate a correlation between the gene dose of Ink4/Arf and resistance to aging-associated cancer. Of relevance, these observations also indicate that the transgenic locus is not subject to aging-associated epigenetic silencing and that both transgenic copies remain active at advanced ages yielding the expected phenotype of increased cancer protection.

image

Figure 3. Enhanced cancer protection and extended longevity in Ink4/Arf-tg/tg mice. (A) Spontaneous cancer incidence in old mice of the indicated genotypes. Mice were sacrificed when they showed overt signs of poor health, such as reduced activity or dramatic weight loss, and analysed for malignant lesions by histology. Statistical significance vs. wt mice was calculated using the Fisher's Exact test: #P < 0.1; ***P < 0.001. (B) Survival curves of cohorts of mice with the indicated Ink4/Arf genotype. Statistical significance of the Kaplan-Meier curves was assessed vs. wt mice using the logrank test: tg mice, P = 0.1; tg/tg mice, P = 0.001. (C) Cancer-free survival curves. The survival data used for the curves in (B) were filtered to exclude those animals that presented malignant tumours at the time of death. Logrank test vs. wt mice: tg mice, P = 0.1; tg/tg mice, P = 0.05. (D) Incidence of kidney lesions in old moribund mice. Most kidney lesions correspond to glomerulonephritis with chronic interstitial nephritis. Statistical analysis was done as in (A). For the wt cohort: a total of 111 mice were scored for survival, 95 of them were analysed by necropsy and histopathology, 49 were free of tumours and the rest, 46, had tumours at the time of death. For the tg cohort: a total of 54 mice were scored for survival, 40 were analysed, 25 were free of tumours and 15 had tumours at the time of death. For the tg/tg cohort: a total of 25 mice were scored, 19 were analysed, 17 were free of tumours and 2 presented tumours. (E) Survival curves of cohorts of mice with the indicated p53 genotype (P > 0.5 in all comparisons).

Download figure to PowerPoint

Increased longevity in Ink4/Arf-tg/tg mice

As previously described (Matheu et al., 2004), Ink4/Arf-tg mice had a normal lifespan, not significantly different from wt mice (log rank test, P = 0.09) (Fig. 3B). Remarkably, however, Ink4/Arf-tg/tg mice presented a significantly different survival curve (logrank test, P = 0.001) with a more square shape. Median survival was extended from 27.5 months to 31.7 months, i.e. an extension of 15% (see detailed data in Supplementary Table S1). To dissociate the possible contribution of decreased spontaneous cancer to the extended lifespan, we eliminated from the previous graph those mice that presented malignant tumours at their time of death, thus focusing on the longevity of cancer-free mice (Fig. 3C). Importantly, cancer-free Ink4/Arf-tg/tg mice still presented a clear increase in median lifespan compared to the corresponding wt mice (Fig. 3C and Supplementary Table S1). This suggests that the Ink4/Arf transgene has an impact on longevity that cannot be attributed to its cancer protection activity. Of note, similar to Ink4/Arf/p53-tg mice (Matheu et al., 2007), Ink4/Arf-tg/tg mice showed no extension in the maximum lifespan (Fig. 3B), suggesting the existence of aging-inducing mechanisms that are not mitigated by the Ink4/Arf locus.

Together with cancer, kidney lesions, mostly consisting in glomerulonephritis with chronic interstitial nephritis, constitute the most frequent pathology in aged C57BL6 mice affecting more than 50% of the wt mice (Hayashi et al., 1988; Lipman et al., 1998; Matheu et al., 2007). Interestingly, old Ink4/Arf-tg/tg showed a significantly lower incidence of kidney lesions compared to wt mice (Fig. 3D).

Finally, to evaluate whether the longevity effect of the Ink4/Arf locus was exclusive of this tumour suppressor or was a general property associated to anti-cancer genes, we performed a similar longevity analysis in mice with increased gene dosage of p53. Before, we had reported that p53-tg mice, carrying one additional transgenic copy of p53, are significantly protected from cancer compared to wt mice (Garcia-Cao et al., 2002). Interestingly, a single p53-tg allele had no impact on normal aging or on aging driven by short telomeres (Garcia-Cao et al., 2002; Garcia-Cao et al., 2006), but in combination with the Ink4/Arf transgenic allele (Ink4/Arf/p53-tg mice) it resulted in an increased median longevity (Matheu et al., 2007). Considering that the longevity effect that we are reporting here is only clearly noticeable in doubly transgenic Ink4/Arf-tg/tg, we considered of importance to perform a longevity study in an equivalent cohort of doubly transgenic p53-tg/tg mice (as before, all cohorts are in a pure C57BL6 genetic background). Interestingly, p53-tg/tg mice had the same longevity profile as p53-tg and wt mice (Fig. 3E). It should be noted that these results do not negate an anti-aging effect of p53, but only indicate that increasing p53 by itself, as in p53-tg/tg mice, is not sufficient to impact longevity. In summary, we conclude that the Ink4/Arf locus possesses an intrinsic capacity to increase longevity that is independent of its cancer protection activity.

Delayed aging in Ink4/Arf-tg/tg mice

We wondered whether the increased median longevity could be correlated with a delayed onset of aging. For this, we examined various biomarkers of aging in old mice of the three relevant genotypes. In particular, we examined aging markers previously validated by others and by us (Sedelnikova et al., 2004; Herbig et al., 2006; Matheu et al., 2007), such as the neuromuscular coordination assay, hair regrowth, and DNA damage signalling in the liver, the latter measured as the percentage of cells positive for γH2AX or 53BP1 (Fig. 4A–D). In all these assays, old Ink4/Arf-tg/tg performed better than wt mice; and in the case of hair regeneration and DNA damage accumulation, Ink4/Arf-tg/tg mice also performed better than singly transgenic mice (Fig. 4A–D). Together, these observations suggest that the Ink4/Arf locus has the capacity to delay aging.

image

Figure 4. Biomarkers of aging in Ink4/Arf-tg/tg mice. (A) Neuromuscular coordination assay, also known as tightrope test, of old (> 20 months) mice (wt, n = 23; tg, n = 26; tg/tg, n = 12) of the indicated genotypes. Statistical significance vs. wt mice was calculated using the Fisher's Exact test: *P < 0.05. (B) Hair re-growth capacity of aged (≥ 24 months) mice (wt male, n = 5; wt female, n = 6; tg male, n = 3; tg female, n = 7; tg/tg male, n = 4; tg/tg female, n = 4). Hair regrowth was scored 15 days after plucking as explained in Methods. Statistical significance vs. wt mice was calculated using the Student's t-test: #P = 0.06). (C and D) Aging-associated accumulation of γH2AX-positive cells (C) and 53BP1-positive cells (D) in the liver. Cryosections from aged (≥ 24 months) mice (n = 3–4 per genotype) were analysed for the percentage of γH2AX positive nuclei by confocal microscopy. Data are mean values ± SEM, Student's t-test is relative to wt: ***P < 0.001.

Download figure to PowerPoint

Antioxidant defences in Ink4/Arf-tg/tg mice

Recent evidence has demonstrated that p53, upon exposure to low intensity stress, such as it might happen during aging, moderately activates anti-oxidant genes, most prominently sestrins Sesn1 and Sesn2 (Sablina et al., 2005). Previously, we have reported that Ink4/Arf/p53-tg mice have higher hepatic expression of Sesn1 and Sesn2, and we have also shown that the livers from these mice have a higher global anti-oxidant potential as reflected by their higher levels of reduced-glutathione (GSH) (Matheu et al., 2007). Importantly, we recapitulated the same observations in old (≥ 24 months) Ink4/Arf-tg/tg mice both regarding the expression of Sesn1 and Sesn2 (Fig. 5A), as well as, with regard to GSH levels (Fig. 5B). Also, the levels of reactive oxygen species in the spleens of Ink4/Arf-tg/tg mice were lower than in Ink4/Arf-tg mice (Supplementary Fig. S2). Together, these findings suggest that Ink4/Arf-tg/tg mice are protected against oxidative damage through an increased expression of anti-oxidant genes probably mediated by p53.

image

Figure 5. Increased anti-oxidant protection in Ink4/Arf-tg/tg mice. (A) Expression of antioxidant genes in vivo. Liver samples from aged (≥ 24 months) mice (n = 3 per genotype) were analysed for the expression of the antioxidant genes sestrin 1 (Sesn1) and sestrin 2 (Sesn2) by real-time quantitative RT-PCR. PCR data were normalized to β-actin expression and are expressed relative to gene expression levels in wt tissues. (B) Levels of reduced glutathione (GSH) in the liver of aged (≥ 24 months) mice (wt, n = 8; tg, n = 7; tg/tg, n = 6). GSH was measured by high-performance liquid chromatography. Data correspond to the mean ± SEM. Statistical significance vs. wt was calculated using the Student's t-test: #P = 0.02; *P < 0.05).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Here, we have addressed the global impact of the entire Ink4a/Arf/Ink4b locus (abbreviated as Ink4/Arf) on fertility, cancer, aging and longevity.

The Ink4/Arf locus is strongly upregulated during human and mouse aging (Zindy et al., 1997; Krishnamurthy et al., 2004; Ressler et al., 2006), and it has been genetically linked to numerous aging-associated human diseases including general organismal frailty [reviewed in Sharpless & DePinho (2007), Melzer (2008); see also Matarin et al. (2008)]. However, these observations do not address whether the expression of the Ink4/Arf locus promotes or protects from aging. Work with genetically modified mice has generated a wealth of valuable information on the role of the Ink4/Arf locus in cancer and stem cell biology, but has not directly examined the impact of the locus on aging. It is interesting to note that the effects of a transgenic allele constitutively overexpressing p16Ink4a (from 2- to 20-fold depending on the tissue), were minimal or absent in the pancreatic islets and in the neural stem cells of old mice (Krishnamurthy et al., 2006; Molofsky et al., 2006). Regarding p19Arf, we and others have concluded that it has an anti-aging effect (Matheu et al., 2007; Baker et al., 2008). Considering these precedents, we regard of relevance to study directly the global impact of the Ink4/Arf locus on aging and longevity.

The first obvious phenotype that we have noticed in the Ink4/Arf-tg/tg mice is the complete sterility of the males. This prompted us to analyze the testis and sperm production of these mice, observing a clear negative impact of the Ink4/Arf-tg allele that results in progressively smaller testes and sperm counts in singly transgenic and doubly transgenic mice compared to wt mice. Histologically, Ink4/Arf-tg/tg mice were completely devoided of spermatogonia and only presented Sertoli cells in the testicular tubules. A number of evidences point to p16Ink4a and p15Ink4b, but not p19Arf, as responsible for this phenotype. First, male sterility has not been reported for a variety of genetic modifications that upregulate p53 (Garcia-Cao et al., 2002; Tyner et al., 2002; Matheu et al., 2004; Mendrysa et al., 2006; Matheu et al., 2007), although there is one exception (Maier et al., 2004). In contrast, impaired spermatogenesis has been reported for numerous mice carrying genetic modifications that result in inhibition of the Cdk4/Cdk2/E2f1 pathway, which is a downstream target of the Ink4 proteins through inhibition of Cdk4 (Gil & Peters, 2006; Kim & Sharpless, 2006). In particular, severe or complete absence of spermatogonia has been reported for Cdk4-null mice (Rane et al., 1999; Tsutsui et al., 1999; Mettus & Rane, 2003), Cdk2-null mice (Berthet et al., 2003; Ortega et al., 2003), Skp2-null mice (due to their abnormally high levels of the Cdk inhibitor p27Kip2) (Fotovati et al., 2006), and E2f1-null mice (Field et al., 1996; Yamasaki et al., 1996). Together, we conclude that male germ cells are particularly sensitive to the inhibition of the cell cycle by the Ink4/Arf locus, probably due to the effect of the p16Ink4a and p15Ink4b proteins, rather than to the effect of p19Arf.

Regarding cancer susceptibility, as anticipated from our previous data with singly transgenic Ink4/Arf-tg mice (Matheu et al., 2004; Matheu et al., 2007), doubly transgenic mice were even less susceptible to aging-associated cancer, which corroborates the well-established cancer protection activity of the Ink4/Arf locus. The most interesting data came from the longevity curves of these mice because there was a clear and statistically significant increase in median longevity (+15%). The survival curve of the Ink4/Arf-tg/tg mice had a more squared shape than in the case of wt mice (Fig. 3B). For example, at 2 years of age, approximately 25% of the wt mice have died whereas none of the Ink4/Arf-tg/tg mice has died at this age; also, despite retaining the same maximal lifespan, only 5% of wt mice reached 33 months of age, whereas 40% of the Ink4/Arf-tg/tg mice were alive at this age (Fig. 3B). The lack of impact on maximum lifespan is suggestive of additional aging processes that are not mitigated by the Ink4/Arf locus, one of them likely being telomere shortening (Tomas-Loba et al., 2008).

At a first approximation, the increased longevity of the Ink4/Arf transgenic mice could be due to a lower incidence of cancer. However, this explanation was not supported by the longevity curves of the cancer-free mice, which still presented a clear increase in longevity. Moreover, despite the potent effect of the transgenic p53 allele on cancer protection (Garcia-Cao et al., 2002; Matheu et al., 2007; Tomas-Loba et al., 2008), mice p53-tg and p53-tg/tg had longevity curves indistinguishable from the control wt cohort, further implying that death by cancer has a small contribution to longevity in C57BL6 laboratory mice. This observation does not negate a role of p53 in aging, but it reinforces the idea that the Ink4/Arf locus possesses a potent anti-aging activity that is separable from its anti-cancer effect.

The analysis of pathologies in moribund aged mice showed a significant difference in the incidence of kidney lesions, which is the most common aging-associated pathology, together with cancer, in C57BL6 mice (Hayashi et al., 1989; Lipman et al., 1998; Matheu et al., 2007). Doubly transgenic Ink4/Arf-tg/tg mice presented an incidence of kidney lesions (26%) that was less than half the incidence observed in the control wt mice (60%). We do not know at present whether this is a direct effect of the Ink4/Arf locus on renal biology or whether this is secondary to other physiological processes, but this beneficial effect on renal function could certainly contribute to the increased longevity of Ink4/Arf transgenic mice. In this context it is interesting to mention the case of mice lacking Skp2, and hence overexpressing the Cdk2 inhibitor p27Kip2. These mice present a lower proliferative response upon severe kidney damage compared to wt mice and, interestingly, this decreased compensatory proliferation is indeed protective and helps to preserve renal function in the face of damage (Suzuki et al., 2007). Based on these data, we speculate that a lower proliferative response in doubly transgenic Ink4/Arf-tg/tg mice may protect renal function at old ages.

In addition to the above-mentioned pathological analyses, we have examined a number of aging-associated markers that allow to evaluate the progression of aging in particular organs or systems. Specifically, aged transgenic mice performed better than wt mice in the tightrope assay that measures neuromuscular coordination, and in the hair re-growth assay. Finally, we examined the presence of molecular markers of aging-associated DNA damage response (Sedelnikova et al., 2004; Herbig et al., 2006; Matheu et al., 2007). When compared to wt mice, the livers of aged Ink4/Arf-tg/tg had a lower percentage of γH2AX and 53BP1 positive cells. These data on aging biomarkers, together with the increased longevity and the lower incidence of aging-associated kidney lesions, indicate that the Ink4/Arf locus contributes to maintain tissular fitness at old ages.

When considering the mechanisms that could explain the anti-aging activity of the locus, it is best to consider p19Arf separately from the Ink4 proteins. In the case of p19Arf, current evidences, based on the combination with an extra gene copy of p53 or with a hypomorphic mutation in BubR1, suggest that p19Arf possesses anti-aging activity (Matheu et al., 2007; Baker et al., 2008). The anti-aging activity of p19Arf is likely mediated by the stabilization of p53, which upregulates the basal expression of anti-oxidant genes and eliminates irreversibly damaged cells (Vousden & Lane, 2007). In support of this, we have observed increased basal levels of the anti-oxidant p53 targets Sestrins and of GSH (reduced-glutathione). However, as shown here, increased p53 has not the same effect on aging as increased Ink4/Arf, despite similar quantitative protection from aging-associated cancer (Matheu et al., 2007). This suggests that p19Arf-mediated activation of p53 conveys an anti-aging activity that is not achieved by simply increasing p53 gene dosage and this probably reflects the ability of the Ink4/Arf locus to respond to mitogenic over-stimulation.

Regarding p16Ink4a, we envision two opposite scenarios: First, Ink4a could be a pro-aging gene whose negative effects on aging are surmounted by the anti-aging activity of Arf. However, as discussed above, current data on transgenic mice overexpressing Ink4a have not found significant effects on stem cell numbers or tissue proliferation at advanced ages (Krishnamurthy et al., 2006; Molofsky et al., 2006). Alternatively, we favour the possibility that Ink4a and Ink4b are indeed anti-aging genes. We propose that the mechanism to explain the anti-aging activity of Ink4a,b is precisely their capacity to decrease proliferation. There are examples in support of the concept that low proliferation rates are beneficial at advanced ages or for the long-term viability of stem cell pools. First, calorie restriction, the gold standard of anti-aging interventions, decreases global proliferation rates [see, for example Varady et al. (2008)]. Also, low levels of the growth factor IGF1 are characteristically associated to aging and play an anti-aging role (Garinis et al. 2008). Finally, ablation of p53 or its target p21Cip1 results in loss of quiescence and depletion of long-term haematopoietic stem cells and neural stem cells at advanced ages (Cheng et al., 2000; Kippin et al., 2005; Liu et al., 2009).

It is conceivable that the upregulation of the Ink4a/Arf locus during aging reflects a protective response elicited by an aberrant or unbalanced mitogenic milieu. In the case of Ink4a,b, their anti-aging activity may be due to their ability to oppose mitogenic over-stimulation, thus avoiding unnecessary exhaustion of proliferative potential, particularly in progenitor and stem cells. In the case of Arf, its anti-aging activity is probably mediated by p53, including its anti-oxidant effect and, also, its ability to decrease proliferation rates. Although the proposed mechanisms are conceptually similar to the mechanisms that explain the anti-cancer activity of the Ink4/Arf locus, it is important to note that they are quantitatively different: progressive upregulation of the locus during aging likely results in slower proliferation rates; whereas acute upregulation of the locus during oncogenesis results in complete cessation of proliferation. Future work will be necessary to dissect the detailed anti-aging mechanisms involved.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Mice

Mice were housed at the pathogen-free barrier area of the Spanish National Cancer Research Center (CNIO), Madrid. Mice were observed on a daily basis and sacrificed when they showed signs of morbidity or overt tumours, in accordance to the Guidelines for Humane Endpoints for Animals Used in Biomedical Research. Doubly transgenic Ink4/Arf-tg/tg mice were generated by intercrossing singly transgenic mice Ink4/Arf-tg (Matheu et al., 2004). The genetic background of all the mice used in this study is > 99% C57BL6 (after at least seven backcrosses with pure C57BL6).

Tightrope test and hair regrowth assay

For the tightrope test, also known as neuromuscular coordination assay, mice were placed on a bar of circular section (60 cm long and 1.5 cm diameter) and the test was considered successful when a mouse did not fall during a period of 60 s in at least one trial out of five consecutive trials. For the hair re-growth assay, dorsal hair was removed by plucking from a square of approximately 1.5 cm × 1.5 cm. Hair re-growth was scored 2 weeks later based on digital photographs and a semi-quantitative assessment using an arbitrary scale from 1 to 4 (four being complete hair regeneration). Scoring was done blindly by two investigators who obtained essentially identical scores.

Sperm count

For each mouse (8-week-old), the cauda epididymidis and vas deferens from each side were harvested separately. The sperm-containing fluid was squeezed out of the vas and the cauda was cut into pieces to extract the remaining sperm. The sperm fluid and the pieces of cauda were suspended in 1 mL of DMEM containing 25 mm Hepes and incubated during 10 min at room temperature to allow the sperm to swim out. The total number of spermatozoa was counted with a haematocytometer for each cauda/vas and the average between the two was obtained for each mouse.

Histological and immunohistochemical methods

For hematoxylin and eosin staining, testes were fixed overnight in Bouin's fixative. For immunohistochemistry, testes were fixed first in formalin for 4 hours and then overnight in diluted Bouin's fixative. After fixation, testes were stored in 70% ethanol at 4 °C until processed. Tissues were dehydrated in increasing concentration of ethanol, embedded in paraffin wax, and sectioned at a thickness of 5 µm. Sections were stained with haematoxylin and eosin or processed for inmunohistochemical analysis with the following antibodies: anti-Ink4a (M156, Santacruz), anti-Arf (5-C3–1, SantaCruz), Ki67 (310QD, Master Diagnostica) and anti-Sox9 (AB5535, Chemicon).

Determination of DNA damage in tissues

DNA damage was assessed by confocal immunofluorescence against 53BP1 (from Novus Biologicals) or γH2AX (antibody clone JBW301 from Upstate Biotechnology) on cryosections as previously described (Garcia-Cao et al., 2006).

Quantitative RT-PCR

Total RNA was extracted from liver with Trizol (Invitrogen, Carlsbad, CA). Reverse transcription was performed using random priming and Superscript Reverse Transcriptase (Life Technologies), according to the manufacturer's guidelines. Quantitative real-time PCR was performed using DNA Master SYBR Green I mix (Applied Biosystems) in an ABI PRISM 7700 thermocycler (Applied Biosystems, Carlsbad, CA). Variations in input RNA were corrected by substracting the number of PCR cycles obtained for β-actin. The primers used were: Sesn1: 5′-CCA GGT AGG AAC ACT GAT GC-3′ and 5′-GTC TGG ATA ACA TCA CAT TAG-3′; Sesn2, 5′-CTC ACA GCT GGT CTG TGT G-3′ and 5′-CCT CCG TGT GGC AAT ACC-3′; Ink4a, 5′-AAC TCT TTC GGT CGT ACC CC-3′ and 5′-GCG TGC TTG AGC TGA AGC TA-3′; Arf 5′-GCC GCA CCG GAA TCC T-3′ and 5′-TTG AGC AGA AGA GCT GCT ACG T-3′; actin, 5′-GGC ACC ACA CCT TCT ACA ATG-3′ and 5′-GTG GTG GTG AAG CTG TAG CC-3′.

Determination of reduced glutathione

Reduced glutathione (GSH) was determined by high-performance liquid chromatography (HPLC) as described (Asensi et al., 1994).

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

We are indebted to Maribel Muñoz for excellent mouse colony management and Marta Riffo for help with sperm extraction. A. Matheu was supported by a predoctoral fellowship from the Spanish Ministry of Education and Science (MEC). A. Maraver is funded by the ‘Juan de la Cierva’ Program (MEC) and MC by the ‘Ramon y Cajal’ Program (MEC). Work at the laboratory of MS is funded by the CNIO and by grants from the MEC (SAF2005-03018 and OncoBIO-CONSOLIDER), from the Government of Madrid (GsSTEM), from the European Union (INTACT and PROTEOMAGE), and from the ‘Marcelino Botin’ Foundation.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Fig. S1 Expression of Ink4a and Arf in old transgenic mice.

(A) Spleen histological sections from aged (≥ 24 months) mice were analysed for the expression of p19Arf by immunohistochemistry. Quantifications indicate average ± SE of positive cells. (B) Liver samples from aged (≥ 24 months) tg (n = 5) or tg/tg (n = 5) mice were analysed for the expression of Ink4a and Arf by quantitative real-time PCR. PCR data were normalized to β-actin expression and are expressed relative to gene expression levels in tg tissues (mean ± SEM.).

Fig. S2 Reactive oxygen species (ROS) in old transgenic mice.

Splenocytes from aged (≥ 24 months) tg (n = 2) or tg/tg (n = 2) samples were analysed for ROS levels by FACS analysis using DCF. Data are mean values ± SEM. Normalized to ROS levels in tg splenocytes, which were set to 100%.

Table S1 Delayed aging of Ink4/Arf transgenic mice.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

FilenameFormatSizeDescription
ACEL_458_sm_Suppmat.pdf197KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.