After a great deal of early promise, resveratrol has been on the ropes for a while, most prominently as a result of studies questioning whether it can directly activate sirtuins — this against a backdrop of growing skepticism that sirtuin activation can extend mammalian lifespan in any case.

Now, another (possible) black eye: GlaxoSmithKline (the company that purchased Sirtris, a pharmaceutical company co-founded by sirtuin/resveratrol pioneer David Sinclair) has suspended a trial of a resveratrol formulation, SRT501 in multiple myeloma patients, because several of the study’s subjects developed kidney failure.

GSK emphasizes that the trial has not been cancelled, but they are observing a moratorium on recruiting new patients until they determine whether the resveratrol was responsible for the subjects’ kidney problems. Nephropathy is a frequent complication in myeloma; one hypothesis being entertained is that the very high doses of resveratrol used in the trial caused vomiting, which in turn resulted in dehydration and tipped the balance in kidneys already close to failure due to the underlying cancer.

More elsewhere:


(previous session)

Matt Hirschey (Verdin Lab, UCSF-Gladstone): Lack of SIRT3 results in the metabolic syndrome. SIRT3 is a mitochondrial sirtuin (NAD+-dependent deacetylase) that is upregulated in liver upon fasting; knockout mice (SIRT3KO) are grossly normal but have trouble with lipid metabolism (specifically, beta-oxidation). Hershey identified several mitochondrial proteins involved in lipid oxidation that are deacetylated in response to fasting, in wildtype but not SIRT3KO. The knockouts are prone to developing obesity and metabolic syndrome with age.

Kate Brown (Chen lab, UC-Berkeley): Calorie restriction reduces oxidative stress by inducing SIRT3. Beginning with an invocation of the free radical theory of aging, and the observation that calorie restriction (CR) reduces oxidative stress, Brown asked whether the mitochondrial sirtuin SIRT3 could be involved in resistance to reactive oxygen species. She showed that CR induces SIRT3 expression, and that the SIRT3 protein deacetylates the mitochondrial antioxidant enzyme SOD2. Furthermore, consistent with Subhash Katewa’s talk in the first session, she demonstrated that CR reduces oxidative stress by switching from glucose to fatty acid oxidation, and that this switch requires SIRT3 activity.

(We’ve discussed SIRT3 before, most recently regarding its role as a tumor suppressor and also with respect to its relationship with exercise).

Ruth Tennen (Chua lab, Stanford): Insight into SIRT6 function at telomeres and beyond. Another member of the sirtuin family, SIRT6, is not localized to mitochondria but rather to telomeres, where it maintains telomeric chromatin in a healthy state and regulates the activity of the senescence-associated transcription factor NF-κB – for more background, see this previous post.) Tennen has shown that SIRT6 is involved in regulating the telomere position effect (TPE) – the silencing of gene expression caused by proximity to a telomere. The TPE has been implicated in age-related changes in gene expression: as telomeres shorten over time, telomere-proximal genes are aberrantly expressed — meanwhile, silencing factors are liberated to wander throughout the genome, repressing genes that should be turned on; similar logic has been applied to the relationship between DNA damage and transcriptional dysregulation.

Jue Lin (Blackburn Lab, UCSF): Telomere length maintenance and aging-related diseases. This talk described work that builds on significant progress, from this lab and others, demonstrating relationships between telomere length and stress, psychological outlook, and lifespan. Lin reviewed evidence that perceived stress is correlated with telomere length in white blood cells (consistent with previous results showing a relationship with intrusive thoughts). New-to-me data included a demonstration that people who increased omega-3 levels or made favorable lifestyle changes exhibited a slower rate of telomere shortening.

(next session)

Yesterday we learned that the most well-characterized mammalian sirtuin, SIRT1, is involved in the control of behavior in response to food availability. SIRT1 is just one of seven sirtuins in mammalian genomes, each of which has a characteristic expression pattern, subcellular localization, and physiological importance.

Today we’re going to talk about another member of the sirtuin family, SIRT3, which has been known for a while now to localize to the mitochondria. Now, Kim et al. have shown that the SIRT3 protein acts as a tumor suppressor:

SIRT3 Is a Mitochondria-Localized Tumor Suppressor Required for Maintenance of Mitochondrial Integrity and Metabolism during Stress

The sirtuin gene family (SIRT) is hypothesized to regulate the aging process and play a role in cellular repair. This work demonstrates that SIRT3−/− mouse embryonic fibroblasts (MEFs) exhibit abnormal mitochondrial physiology as well as increases in stress-induced superoxide levels and genomic instability. Expression of a single oncogene (Myc or Ras) in SIRT3−/− MEFs results in in vitro transformation and altered intracellular metabolism. Superoxide dismutase prevents transformation by a single oncogene in SIRT3−/− MEFs and reverses the tumor-permissive phenotype as well as stress-induced genomic instability. In addition, SIRT3−/− mice develop ER/PR-positive mammary tumors. Finally, human breast and other human cancer specimens exhibit reduced SIRT3 levels. These results identify SIRT3 as a genomically expressed, mitochondria-localized tumor suppressor.

So, the absence of the SIRT3 gene disrupts mitochondrial function and destabilizes the mitochondrial genome, but the causal relationship is unclear: one can imagine a derangement of mitochondrial morphology or physiology causing the genomic instability, but one can also imagine a primary defect in DNA metabolism causing the physiological defects — and one could also imagine both, operating in a vicious cycle.

The significance of the “single oncogene” observation requires a bit of explanation: In a normal cell, simply turning on a single tumor-promoting gene isn’t sufficient to transform the cell, i.e., create a tumor — if a cell detects a hyperphysiological level of a growth factor or internally generated mitogenic signal, it will undergo senescence, a permanent cell cycle arrest that (among other things) prevents mutated cells from turning into cancers.

In the case of the SIRT3 knockout, however, a single oncogene is enough to transform an otherwise normal cell, i.e., the loss of SIRT3 function appears to potentiate the transformation process. Furthermore, this implies that the mitochondria are involved in the decision to undergo senescence. This is not to say that SIRT3 is directly involved in the senescence fate decision: the SIRT3 knockout has broadly deranged mitochondria, and it’s not clear which mitochondrial function is involved in senescence.

While this report does not fully elucidate the mechanism of SIRT3 action, it is clear that oxidation must be central to the issue: reactive oxygen species (ROS) levels are high in the SIRT3 knockout, but increasing the dose of an antioxidant enzyme prevents the single-oncogene transformation. We don’t yet know whether the ROS themselves are causing oncogenic mutations, or the loss of the mitochondrial function in senescence is allowing cells that would have arrested to progress further down the path to cancer. (Could be both, obviously, but it’s likely that one of the two is more important.)

These findings are consistent with what we already know about the connection between mitochondrial ROS, cellular damage and lifespan (ideas that are central to one strategy for developing longevity-enhancing drug compounds by targeting antioxidants to mitochondria), but they raise a question: how much of the “longevity regulation” we owe to antioxidant genes can be attributed to tumor suppression?

ResearchBlogging.orgKim, H., Patel, K., Muldoon-Jacobs, K., Bisht, K., Aykin-Burns, N., Pennington, J., van der Meer, R., Nguyen, P., Savage, J., & Owens, K. (2010). SIRT3 Is a Mitochondria-Localized Tumor Suppressor Required for Maintenance of Mitochondrial Integrity and Metabolism during Stress Cancer Cell, 17 (1), 41-52 DOI: 10.1016/j.ccr.2009.11.023

There’s lots more about SIRT3 in a recent post at @ging.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Not only does the mammalian sirtuin SIRT1 mediate the lifespan extension phenotype of caloric restriction (CR), it is also involved in controlling behavior (such as food intake) in response to CR (and possibly during ad libitum feeding as well).

Two recent papers with consistent results address the issue. Both studies employed brain-specific knockouts of SIRT1; Cohen et al. used a brain-specific knockout, whereas Çakir et al. used both pharmacologic inhibition and an siRNA in the hypothalamus. The latter paper implicates the FoxO1 transcription factor and S6 kinase signaling, implying cross-talk with both the IGF-1 and TOR pathways.

ResearchBlogging.orgÇakir, I., Perello, M., Lansari, O., Messier, N., Vaslet, C., & Nillni, E. (2009). Hypothalamic Sirt1 Regulates Food Intake in a Rodent Model System PLoS ONE, 4 (12) DOI: 10.1371/journal.pone.0008322

Cohen, D., Supinski, A., Bonkowski, M., Donmez, G., & Guarente, L. (2009). Neuronal SIRT1 regulates endocrine and behavioral responses to calorie restriction Genes & Development, 23 (24), 2812-2817 DOI: 10.1101/gad.1839209

Cells tend to produce unwanted protein aggregates and other molecular refuse slightly faster than they can get rid of it, resulting in a time-dependent accumulation of potentially toxic cellular garbage. This, in turn, can cause an age-dependent loss of cellular viability, which is (in certain contexts) a fair operational definition of aging.

How can cells deal with their garbage? Protein aggregates are both sticky and insoluble, making it hard for cellular machinery to deal with them at an enzymatic level. If the gunk can’t be eliminated, however, it might still be possible to move it around in a useful way. Specifically, at mitosis, the cell could make sure that all the potentially toxic aggregates stay in one of the progeny. To illustrate the argument I’ll turn to the words of the estimable Alex Palazzo:

One approach is to distribute everything equally amongst your two offspring. …

A second approach is to give all the crap to one of the two new cells and keep the other one pristine. Lets call these two cells the crap cell and the pristine cell. What’s the result of this second strategy? Using our crap metric from above, the first cell accumulates 10 units of garbage over its lifetime and then gives it all to one offspring, the crap cell, and none to the other offspring, the pristine cell. Those cells then grow and by the time they divide each second generation cells have made 10 units of additional crap each. The crap cell has 20 units the pristine cell 10. The two cells divide and dump all their garbage on one of their offsprings. One cell starts with 20 units of crap, one cell with 10 units and two cells are again crap free. The end result of this strategy? Part of your descendents will become more and more decrepit as they fill up with crap, while others remain pristine.

The crap cell (I love this nomenclature) will become inviable sooner under this strategy, but the alternative would be a symmetric division strategy in which all descendants accumulate garbage, ultimately causing the extinction of the entire lineage. The idea here is that assuming certain values for adjustable parameters re: the rate of garbage accumulation and the effect of garbage level on reproductive fitness, this can be an advantageous strategy to ensure reproductive success. Both single-celled yeast and mammalian stem cells employ this asymmetric strategy in order to preserve the viability of an indefinitely dividing lineage.

In yeast, the crap cell is called the “mother”; the pristine cell is called the “daughter” — mom accumulates garbage of various kinds, both protein aggregates and rDNA circles. When the mother is ready to divide, a bud forms at a specific site on her cell wall, defined by a set of macromolecular complexes that determine cellular polarity. Liu et al. have demonstrated that the daughter cell is using some of the same polarity-determining machinery (the “polarisome”) to actively transport protein aggregates back into the mother:

The Polarisome Is Required for Segregation and Retrograde Transport of Protein Aggregates

The paradigm sirtuin, Sir2p, of budding yeast is required for establishing cellular age asymmetry, which includes the retention of damaged and aggregated proteins in mother cells. By establishing the global genetic interaction network of SIR2 we identified the polarisome, the formin Bni1p, and myosin motor protein Myo2p as essential components of the machinery segregating protein aggregates during mitotic cytokinesis. Moreover, we found that daughter cells can clear themselves of damage by a polarisome- and tropomyosin-dependent polarized flow of aggregates into the mother cell compartment. The role of Sir2p in cytoskeletal functions and polarity is linked to the CCT chaperonin in sir2Δ cells being compromised in folding actin. We discuss the findings in view of recent models hypothesizing that polarity may have evolved to avoid clonal senescence by establishing an aging (soma-like) and rejuvenated (germ-like) lineage.

Note the role for Sir2p, the founding member of the sirtuin family of longevity assurance genes: Sir2p is required, via another protein’s activity, for the normal folding of actin, the cytoskeletal protein from which the daughter-mother transport cable is built. It’s an indirect interaction, and more complex than I’m making it out to be here. Nonetheless, it is satisfying for those of us looking for unifying theories in aging that one of the most widely studied proteins in lifespan regulation is involved in the deep connection between polarity and aging.

I’ll close with a few questions:

  • Why can’t the mother cell export the aggregates? One of our initial premises was that aggregates are biochemically hard to handle, which is why they accumulate rather than being degraded. But now we know that cells can bundle aggregates onto actin cables and move them around — why not sort the aggregates into vesicles or membrane blebs and dispose of them? Granted, in order to export an aggregate out of the cell, it would have to cross a membrane, but this would be no more difficult topologically than mitophagy. The obvious (and trivial) answer to this question is “because it didn’t evolve that way,” but I’m curious to know whether there’s some compelling reason why it couldn’t have evolved that way.
  • How do symmetrically dividing cells overcome this problem? In order to exploit asymmetric division, one must first establish polarity. The argument above about the rate of garbage accumulation would seem to apply equally well to non-polarized cells like bacteria – why, then, do clonal lineages of symmetrically dividing cells not invariably go extinct? Maybe the cells that we think are symmetric are secretly asymmetric, with a crap/pristine segregation that has yet to be uncovered. Or maybe the symmetric cells know something about garbage disposal that we don’t. In either case, there’s something important to learn that might help us keep mammalian cells youthful.

ResearchBlogging.orgLiu, B., Larsson, L., Caballero, A., Hao, X., Öling, D., Grantham, J., & Nyström, T. (2010). The Polarisome Is Required for Segregation and Retrograde Transport of Protein Aggregates Cell, 140 (2), 257-267 DOI: 10.1016/j.cell.2009.12.031

A review I should have cited yesterday addresses the promise and challenges of the two most prominent natural-product candidates for longevity-enhancing therapeutics. The author is prominent biogerontologist and all-around bright feller Matt Kaeberlein (see here for earlier posts on his group’s work).

Resveratrol and rapamycin: are they anti-aging drugs?

Studies of the basic biology of aging have advanced to the point where anti-aging interventions, identified from experiments in model organisms, are beginning to be tested in people. Resveratrol and rapamycin, two compounds that target conserved longevity pathways and may mimic some aspects of dietary restriction, represent the first such interventions. Both compounds have been reported to slow aging in yeast and invertebrate species, and rapamycin has also recently been found to increase life span in rodents. In addition, both compounds also show impressive effects in rodent models of age-associated diseases. Clinical trials are underway to assess whether resveratrol is useful as an anti-cancer treatment, and rapamycin is already approved for use in human patients. Compounds such as these, identified from longevity studies in model organisms, hold great promise as therapies to target multiple age-related diseases by modulating the molecular causes of aging.

Note that resveratrol has been taking a bit of a thrashing of late, with recently released studies calling into question its ability to directly activate sirtuins. Briefly, the critics posit that the early data may have been misinterpreted due to artifacts in a fluorescence-based system used to detect protein-drug interactions — but check comment #32 on that post for David Sinclair’s personal response on this issue.

ResearchBlogging.orgKaeberlein, M. (2010). Resveratrol and rapamycin: are they anti-aging drugs? BioEssays, 32 (2), 96-99 DOI: 10.1002/bies.200900171

The idea that translation fidelity might play a role in aging dates back at least as far as 1963, when Leslie Orgel proposed the “error catastrophe” theory of aging: in this model, mistranslation of the translational machinery creates a feedback loop that leads to further translation errors, ultimately causing loss of cell viability. From the Science of Aging Timeline:

Orgel considers two types of proteins: those involved in metabolism, and those involved in information processing. For metabolic proteins, translational error isn’t a long-term problem for the cell, since a malfunctioning protein is simply one of many. Likewise, for translational errors causing loss of function in information processing proteins: the error isn’t heritable, and a small decrease in the efficiency of gene expression is unlikely to pose a serious problem.

However, information processing proteins can be altered in another way: by mutations that decrease the fidelity with which they process or propagate genetic information. Lower-fidelity transcription and translation will result in more mutations. This is the core of Orgel’s idea: “errors which lead to a reduced specificity of an information-handling enzyme lead to an increasing error frequency. Such processes are clearly cumulative and…in the absence of an imposed selection for “accurate” protein-synthesizing units, must lead ultimately to an error catastrophe; that is, the error frequency must reach a value at which one of the processes necessary for the existence of viable cell becomes critically inefficient.”

The logic of the feedback loop is compelling, but the theory suffered for lack of experimental verification. While there is still some controversy over whether error catastrophe has received a full and fair experimental test, the consensus appears to be that while error catastrophe can take place under some systems (e.g., viral replication in the presence of drugs that reduce polymerase fidelity), this phenomenon does not play a role in mammalian aging: the measured values of the relevant parameters (basal translation error rates; the likelihood that a given error will result in further alteration to translation fidelity; protein lifetimes; etc.) appear to be such that the feedback loop doesn’t actually occur.

The error catastrophe theory is still an important waypoint in the evolution of theories of aging, and it has had tremendous influence in other areas within biogerontology. For example, similar logic has been applied to the role of autophagy in aging, where the feedback loop is called the garbage catastrophe.

And even if the feedback-loop logic doesn’t hold up to experimental scrutiny, recent findings have revealed that there may nonetheless be a relationship between protein translation fidelity and aging. Writing in PLoS ONE, Silva et al. report that in yeast, increasing the rate of translation errors might increase the activity of the longevity assurance gene SIR2:

The Yeast PNC1 Longevity Gene Is Up-Regulated by mRNA Mistranslation

Translation fidelity is critical for protein synthesis and to ensure correct cell functioning. Mutations in the protein synthesis machinery or environmental factors that increase synthesis of mistranslated proteins result in cell death and degeneration and are associated with neurodegenerative diseases, cancer and with an increasing number of mitochondrial disorders. Remarkably, mRNA mistranslation plays critical roles in the evolution of the genetic code, can be beneficial under stress conditions in yeast and in Escherichia coli and is an important source of peptides for MHC class I complex in dendritic cells. Despite this, its biology has been overlooked over the years due to technical difficulties in its detection and quantification. In order to shed new light on the biological relevance of mistranslation we have generated codon misreading in Saccharomyces cerevisiae using drugs and tRNA engineering methodologies. Surprisingly, such mistranslation up-regulated the longevity gene PNC1. Similar results were also obtained in cells grown in the presence of amino acid analogues that promote protein misfolding. The overall data showed that PNC1 is a biomarker of mRNA mistranslation and protein misfolding and that PNC1-GFP fusions can be used to monitor these two important biological phenomena in vivo in an easy manner, thus opening new avenues to understand their biological relevance.

PNC1 is a longevity gene because its biochemical activity feeds into the sirtuin pathway: Pnc1p synthesizes nicotinic acid from nicotinamide, which is an inhibitor of Sir2p, one of the canonical longevity factors in S. cerevisiae. Overexpression of PNC1 increases lifespan, presumably by increasing the activity of Sir2p. (The authors show that Sir2p silencing activity is elevated under conditions that cause mistranslation, and that this is inhibited by exogenous nicotinamide. Missing, as far as I can tell, is the same experiment in ∆pnc1 cells, which according to the authors’ model would not induce silencing during mistranslation.)

Is this simply an example of a general stressor activating a general stress response, whose constitutive activation in turn makes cells more stress-resistant and therefore longer-lived? For example, one could imagine a translation fidelity problem resulting in synthesis of lots of poorly folded proteins, leading to activation of the heat shock response and expression of chaperones (indeed, in the worm, heat shock transcription factor HSF-1 is required for life extension by daf-2 mutations). This doesn’t appear to be that. Instead, loss of protein fidelity causes upregulation of a major longevity assurance pathway, which acts primarily at the level of transcriptional silencing.

A couple of questions:

  • What is the relevant molecular correlate of translation infidelity? Unfolded proteins would be the most likely culprit (prediction: whether or not it’s involved in the lifespan extension, there should be some heat shock response under these conditions), but one can imagine more elaborate scenarios: Suppose an inhibitor of PNC1 translation is encoded by an mRNA that is particularly likely to be mistranslated under normal conditions (e.g., because of weird codon usage, secondary structure, or some other quirk) and is now translated so poorly that it loses its inhibitory activity altogether (or acquires a new activity).
  • How is the translational upregulation of PNC1 mediated? This is particularly curious given that, by assumption, a cell with a high rate of translation infidelity is having difficulty with translation. Teleologically, there’s no reason not to regulate gene expression at this level — if the gene were upregulated transcriptionally, the mRNA would still have to be translated — but it still strikes me as odd. If this is a bona fide evolved response to translation problems, wouldn’t it be better to pre-synthesize PNC1 and then activate it post-translationally (e.g. by proteolysis)?
  • Is SIR2 involved in translation fidelity? Looking at this story as a straightforward stress response, one would expect some action of SIR2 to help mitigate the stress that started the whole process. So I’d be curious to know whether SIR2 mutants have lower translation fidelity, and if so, how it is that SIR2 is involved in improving the accuracy of translation?

ResearchBlogging.orgSilva, R., Duarte, I., Paredes, J., Lima-Costa, T., Perrot, M., Boucherie, H., Goodfellow, B., Gomes, A., Mateus, D., Moura, G., & Santos, M. (2009). The Yeast PNC1 Longevity Gene Is Up-Regulated by mRNA Mistranslation PLoS ONE, 4 (4) DOI: 10.1371/journal.pone.0005212

Next Page »