Taguchi A, Wartschow LM, White MF.
Brain IRS2 signaling coordinates life span and nutrient homeostasis.
Science. 2007 Jul 20;317(5836):369-72.
Please login to recommend the paper.
To make a comment you must login or register.
Taguchi and colleagues’ recent studies suggest that reducing insulin signaling through deletion of insulin receptor substrate 2 in brain (bIRS2) increases longevity even though it induces increased peripheral insulin resistance and hyperinsulinemia. Neuronal IRS2 is known to mediate critical aspects of systemic energy homeostasis (1); thus, the peripheral metabolic derangement caused by genetic deletion of bIRS2 is not surprising. It is of interest, however, that the typical life-shortening effects of peripheral insulin resistance can be reversed by inactivation of bIRS2-mediated insulin signaling. The authors make a good case that retention of youthful fat and carbohydrate metabolism and postprandial superoxide dismutase response may contribute to the resilience of bIRS2 -/- and +/- animals.
There are mixed implications of these results for understanding the potential contribution of insulin signaling abnormalities to Alzheimer disease (AD) pathogenesis. The finding that brain size is reduced in the bIRS2 -/- both supports the role of IRS2 in brain development and raises the question of potential behavioral deficits in these animals. Further behavioral testing with this model will provide useful data in that regard. There are additional direct implications of this work for Aβ regulation, given suggestions that IRS2 is required for insulin/IGF1-mediated switching from TrkA to the p75NTR neurotrophin receptor, which then stabilizes BACE1 and increases Aβ generation (2). Reducing IRS2-mediated insulin signaling may thus conceivably reduce brain amyloid levels.
One final caveat is needed regarding the extreme plasticity of the ancient and complex insulin signaling pathway. Knockout models of insulin/IGF1 signaling are highly susceptible to effects of reorganization that may not always be easy to discern and control. Replicative findings with conditional knockout models or techniques such as siRNA will help solidify the relevance of these interesting results for human aging and AD.
Comment by Kazuhiro Nakamura and Kun Ping Lu
Aging, Cancer, and Neurodegeneration
Matheu et al. (1) have made the important discovery that overexpression of both Arf and p53 under normal regulation can confer cancer resistance, reduce aging-associated damage, and delay normal aging in mice. These results are especially interesting because these authors have previously shown that overexpression of either Arf or p53 alone can confer cancer resistance, but not longevity (2,3), highlighting the tight regulation of the aging process. The authors have provided a rationale for the co-evolution of cancer resistance and longevity, suggesting that it may be possible to live longer without worrying about cancer.
These results have a general impact on many age-related disorders, including neurodegeneration, which also result from age-related cellular damage in the nervous system. Interestingly, p53-mediated cell death has been associated with the progressive neuronal death in Huntington disease, Parkinson disease, Alzheimer disease, and amyotrophic lateral sclerosis (4). For instance, p53 mediates cellular dysfunction and behavioral abnormalities in Huntington disease (5). In addition, increased expression of Arf and p16ink4a has been shown to decrease brain stem/progenitor cells and neurogenesis during aging and in Bmi1-deficient mice (6-9). Reduced stem/progenitor cell function may be a major cause of the decline in regenerative capacity and contribute to degenerative diseases observed in aging tissues. Thus, it would be of interest to examine whether this Arf/p53 longevity pathway would affect the development of stem/progenitor and progression of neurodegeneration.
The concept of the co-evolution of cancer resistance and longevity is intriguing and significant. However, there are many other examples where longevity and cancer resistance may be antagonistic (10). For example, overexpression of truncated (constitutively active?) p53 leads to premature aging and cancer resistance in mice (11,12). Similarly, we and others have shown that ablation of the prolyl isomerase Pin1 in mice leads to many premature aging phenotypes including neurodegeneration, and also protects against cancer development even induced by overexpression of certain oncogenes or ablation of p53 (13-18). These results are relevant because Pin1 appears to regulate p53 function in response to genotoxic stress (19-21). Therefore, it would be extremely interesting and important to understand how and why cancer resistance and longevity are antagonistic under some conditions, but compatible under some other conditions, and also to investigate the relationship among aging, cancer, and neurodegeneration.
It is well established that p53 is an oncogene, many mutations on which are responsible for the development of various types of cancer. It has been proposed that this pathology is due to the impairment of the ability of p53 to eliminate damaged cells. Aging also results from damaged cells which accumulate, and it is therefore tempting to postulate that p53 could be an endogenous “life prolongator.” Accordingly, in C. elegans, mutations which increase longevity also confer cancer resistance likely via p53 activation. Therefore, longevity and cancer resistance could have in common the potent ability of p53 to clear damaged cells.
The paper by Matheu and colleagues very interestingly documents that in mice, overexpression of p53 and Arf (a p53 stabilizer) triggers cancer resistance and decreases age-associated damage. Therefore, the study gives support to the idea that the control of p53 could be central in aging and provides a rationale to the unexplained observation that long lifetime is apparently associated to increased cancer resistance.
This very interesting and well-performed study raises a few questions. Particularly striking is the observation that the coexpression of both Arf and p53 slightly increases the median lifespan (by 16 percent) but that the shapes of the two survival/age curves are different (see Fig. 2a,b). Although the Arf/p53 mice begin to die later than wild-type mice, the decay in survival seems accelerated for the former mice and finally, both curves gather to give 100 percent mortality at about 150 weeks. In other words, there is not a subset of Arf/p53 mice which survives the group of wild-type mice. Whether the cooperative function of Arf and p53 is efficient at early stages of “old age” but remains ineffective at later stages is a matter of reflection.
Has the work by Matheu implications for the understanding of neurodegenerative diseases? In Parkinson disease-affected brains, an increased p53-dependent cell death has been well documented. In both Parkinson and Alzheimer diseases, proteins responsible for familial cases, such as presenilins or α-synuclein, control p53-dependent cell death, and this function is increased by pathogenic mutations. Even if the demonstration that exacerbated p53-dependent apoptosis is causal in early death observed in genetic cases of AD or PD remains to be established, increased p53 expression accompanying early death would appear paradoxical with respect to Matheu’s work indicating that p53 overexpression triggers enhanced longevity. Close examination of data gives simple explanations. Thus, Matheu and colleagues show that the p53 or Arf overexpression alone did not mimic the phenotype triggered by coexpression of the two protein partners. No data are yet available concerning the levels of Arf in AD or PD brains and whether decreased levels of Arf could account for altered p53-dependent control of the longevity bestowed by p53. It should be noted that if this is the case, such alteration should be topologically restricted to cerebral areas lesioned in these pathologies (which are sometimes very narrow) in contrast to more widely distributed brain zones affected during normal aging. It is therefore more likely that the very nicely documented functional collaboration between Arf and p53 in the control of lifespan is not a central alteration responsible for neurodegenerative diseases associated with exacerbated apoptotic cell death.