Apoptosis


Here is the next in what will likely be a long series of semi-regular review roundups — links, without extensive further comment, to the reviews I found most intriguing over the past few weeks months (I went on hiatus during the winter holidays). For the previous foray into the secondary literature, see here.

Alzheimer’s:

Apoptosis & cancer:

Calorie restriction:

Diabetes:

Klotho:

Sirtuins:

Stem cells:

Telomeres:

Evolution generally rewards early reproduction, but there is some advantage to maintaining reproductive capacity into later life — e.g., if a young organism encounters conditions that would be adverse for offspring viability, it makes sense to delay reproduction or at least to have the capacity to try again when that organism is older. How best to maintain the viability of gametes over the course of the lifespan? The answer appears to be stringent quality control: Andux and Ellis report that in C. elegans, apoptosis in the germ line helps guarantee that high-viability oocytes keep getting the resources they need:

Apoptosis Maintains Oocyte Quality in Aging Caenorhabditis elegans Females

In women, oocytes arrest development at the end of prophase of meiosis I and remain quiescent for years. Over time, the quality and quantity of these oocytes decreases, resulting in fewer pregnancies and an increased occurrence of birth defects. We used the nematode Caenorhabditis elegans to study how oocyte quality is regulated during aging. To assay quality, we determine the fraction of oocytes that produce viable eggs after fertilization. Our results show that oocyte quality declines in aging nematodes, as in humans. This decline affects oocytes arrested in late prophase, waiting for a signal to mature, and also oocytes that develop later in life. Furthermore, mutations that block all cell deaths result in a severe, early decline in oocyte quality, and this effect increases with age. However, mutations that block only somatic cell deaths or DNA-damage–induced deaths do not lower oocyte quality. Two lines of evidence imply that most developmentally programmed germ cell deaths promote the proper allocation of resources among oocytes, rather than eliminate oocytes with damaged chromosomes. First, oocyte quality is lowered by mutations that do not prevent germ cell deaths but do block the engulfment and recycling of cell corpses. Second, the decrease in quality caused by apoptosis mutants is mirrored by a decrease in the size of many mature oocytes. We conclude that competition for resources is a serious problem in aging germ lines, and that apoptosis helps alleviate this problem.

The budding yeast Saccharomyces cerevisiae has been a valuable model system in biogerontology, dating back to the very earliest years of the modern synthesis of molecular genetics with the study of lifespan regulation. From yeast we first learned about the sirtuins, and it continues to teach us much about the mechanisms of lifespan extension by calorie restriction.

Aging in yeast can be studied in one of two ways: by focusing on the replicative lifespan (RLS), which is the number of times that a mother cell can bud to form a daughter; or on the chronological lifespan (CLS), which has to do with how long a cell can persist and maintain viability in a nondividing state. While there is some overlap between the genes regulating CLS and RLS, they are generally discussed as though they are distinct phenomena.

An under-appreciated feature of yeast aging is that, at the end of CLS or RLS, a yeast cell can die either by necrosis or by programmed cell death — i.e., apoptosis or something very much like it. That comes as a surprise to those of us who grew up thinking of apoptosis as a kind of “noble sacrifice” made by a damaged cell in the context of a tissue or organ: damage leads to cancer, but not if it leads to cell death first; hence, there’s a survival benefit to the organism if individual cells “voluntarily” die in response to certain types of stress. But with no body to protect, why would a single-celled organism undergo apoptosis?

The mechanisms and evolutionary ramifications of yeast apoptosis are the subject of a review by Rockenfeller and Madeo. For those of you who have followed this story for a while, Frank Madeo was the first author of the paper that identified caspase-like enzymes operating in yeast apoptosis; that manuscript was a worldwide journal-club favorite throughout the yeast and apoptosis fields back in the early years of the 21st century.

Apoptotic death of ageing yeast

Yeast has been a valuable model to study replicative and chronological ageing processes. Replicative ageing is defined by the number of daughter cells a mother can give birth to and hence reflects the ageing situation in proliferating cells, whereas chronological ageing is widely accepted as a model for postmitotic tissue ageing. Since both ageing forms end in yeast programmed death (necrotic and apoptotic), and abrogation of cell death by deletion of the apoptotic machinery or diminishment of oxidative radicals leads to longevity, apoptosis and ageing seem closely connected. This review focuses on ageing as a physiological way to induce yeast apoptosis, which unexpectedly defines apoptosis as a pro- and not an anti-ageing mechanism.

I take issue with the last sentence in the abstract, at least as it applies to the broader field of biogerontology. Very few of us in the mammalian aging field think of apoptosis as an “anti-aging” mechanism; rather, we see it as an tumor suppressor mechanism that has negative consequences on regenerative capacity. In other words, apoptosis in adult metazoans is an anti-cancer but pro-aging phenomenon.

Attentive readers will notice that I’ve skipped a couple of sessions: Session III was “oral presentations from abstracts,” a series of unrelated short talks; Session IV was a poster session, and Session V was a series of talks about mitochondria that I watched from the bar, where taking out my laptop might have risked a catastrophic beer spill, so I didn’t blog it.

This morning is a session on subjects near and dear to my own heart.

John Sedivy kicked off the series with a discussion of assays for detecting cellular senescence in vivo and in vitro. This is an incredibly important subject, since the current assays have a lot of problems, ranging from poor quantifiability to frank irreproducibility. The Sedivy lab has developed a large number of image-processing protocols that will allow reliable detection and quantitation in multiple system. These techniques will hopefully allow us to nail down once and for all the location, origins, fate, and function of senescent cells in aging tissues. (Heidi Scrable said during questions that her lab has developed a way to do immunohistochemistry after SA ß-gal, a twitchy technique that usually ruins a sample for subsequent analysis; I will definitely be going to that poster this afternoon.)

Next up was Darren Baker, who is doing genetics with progeroid mice: In the BubR1 progeria model, loss of p19 or p53 results in accelerated aging, implying that p53 is involved in delaying aging in vivo. This directly contradicts the idea that p53-induced senescence is a cause of aging, and enters the fray on the side of scholars who believe that properly regulated p53 has a primarily anti-aging function.

Alex Bazarov asked whether p16-induced senescence is reversible in breast cancer cells (it isn’t), and proposed using a small molecule inducer of p16 arrest as a cancer therapy.

Oliver Bischof demonstrated that repetitive DNA is transcribed at the onset of senescence, generating a population of small noncoding RNAs that are sufficient to induce both senescence-associated heterochromatic foci (SAHFs) as well as the senescence growth arrest itself. Note that these small RNAs are distinct from micro-RNAs, whose role in senescence and cancer was the subject of several posters at this meeting (including one by me).

Kan Cao studies progerin, the derivative of Lamin A that is responsible for Hutchinson-Gilford Progeria Syndrome (HGPS). Today’s talk focuses on the role of progerin in normal aging: Normal cells express progressively more progerin as a function of age, but telomerase-immortalized cells express hardly any. Thus, there may be a synergy between telomere shortening and progerin induction during cellular senescence.

Norm Sharpless shared human genetic evidence that variations in the p16/INK4a locus are associated with variations in the rate of aging, cancer, and other age-related diseases (specifically atherosclerosis). The overall results suggest that p16 has diverse effects in different tissues.

Vera Gorbunova discussed the distinct tumor suppressor mechanisms that have evolved in rodents of varying body size and lifespans. She began by introducing the negative correlation between telomerase activity and body size, and between in vitro replicative lifespan and body size. Larger body size means more cell divisions and a greater cancer risk; hence replicative senescence is more common among larger rodents. Another sort of control is observed in naked mole rats, which are long-lived and whose cells exhibit multiple forms of contact-mediated growth arrest. I especially enjoyed the talk because of my recently stoked interest in comparative biogerontology.

On to other exotic organisms: Fish! Our finned friends are getting into the biogerontological act — it was only a matter of time. Shuji Kishi talked about genetic screens in zebrafish that identified mutants showing alterations in senescence-associated biomarkers (specifically, the senescence-associated beta-galactosidase, aka SA ß-gal). One of the mutants he described is deficient in telomere maintenance, and exhibits segmental progeria as well as shortened lifespan; another mutant causes accumulation of lipofuscins, suggesting a defect in lysosomal metabolism or autophagy.

Valery Krizhanovsky, who works right here at Cold Spring Harbor in Scott Lowe’s lab, described a useful function for cellular senescence beyond its well-documented tumor-suppressor function: prevention of liver fibrosis. Senescent cells are present in fibrotic liver in wildtype animals, but in cell-specific p53 knockouts these senescent cells are missing. The senescent hepatic cells appear to attract immune infiltration, which work to clear the senescent cells and in the process alleviate the fibrosis.

Francis Rodier (from the Campisi lab, where I also work) presented his work on the relationship between persistent DNA damage, senescence growth arrest, and the senescence-associated secretory phenotype (SASP). He focused on the regulation of the SASP by an upstream kinase in the DNA damage response pathway — a seminal example of the connection between the chromatin lesions in a compromised genome and the regulation of cell-cell communication.

Lunchtime!

Session index:

AMP-activated kinase (AMPK) agonists mimic the effects of exercise, raising the possibility of a “workout pill” that could simulate the effects of vigorous activity. The applications to human health are, to mildly understate the case, significant; it sounds almost too good to be true, and it leaves one looking for the catch.

But it turns out that AMPK is activated by certain types of genotoxic stress, and contributes to UV-induced apoptosis in the skin. From Cao et al.:

AMP-activated protein kinase contributes to UV- and H2O2-induced apoptosis in human skin keratinocytes

AMP-activated protein kinase or AMPK is an evolutionarily conserved sensor of cellular energy status, activated by a variety of cellular stresses that deplete ATP. However, the possible involvement of AMPK in UV- and H2O2-induced oxidative stresses that lead to skin aging or skin cancer has not been fully studied. We demonstrated for the first time that UV and H2O2 induce AMPK activation (Thr172 phosphorylation) in cultured human skin keratinocytes. UV and H2O2 also phosphorylate LKB1, an upstream signal of AMPK, in an EGFR dependent manner. … We also observed that AMPK serves as a negative feedback signal against UV-induced mTOR (mammalian target of rapamycin) activation in a TSC2 dependent manner. Inhibiting mTOR and positively regulating p53 and p38 might contribute to AMPK’s pro-apoptotic effect on UV- or H2O2-treated cells. Furthermore, activation of AMPK also phosphorylates acetyl-CoA carboxylase or ACC, the pivotal enzyme of fatty acid synthesis, and PFK2, the key protein of glycolysis in UV-radiated cells. Collectively, we conclude that AMPK contributes to UV- and H2O2-induced apoptosis via multiple mechanisms in human skin keratinocytes and AMPK plays important roles in UV-induced signal transduction ultimately leading to skin photoaging and even skin cancer.

Note especially that last line (emphasis mine): activation of AMPK could exacerbate the pro-aging effects that UV light exerts on the skin. Judging from the peroxide results, this also applies to endogenously generated reactive oxygen species (ROS) — which one can’t avoid by simply staying out of the sun.

Before we panic and throw the exercise mimetic baby out with its gerontogenic bathwater, I’d want to see whether AMPK agonists like AICAR do in fact synergize with stresses like UV and peroxide to increase apoptotic cell death in the skin. If they do…well, I think we found that catch.

SIRT1, the most widely studied of the protein family known as sirtuins, is a histone deacetylase that has been implicated in regulation of aging in mammals. Activators of SIRT1, such as resveratrol, have been demonstrated to extend the lifespan as well as boost mitochondrial function in mice.

More recently, SIRT1 has been demonstrated to regulate p53 function: deacetylation by SIRT1 makes p53 less active, thereby decreasing apoptosis in response to specific types of DNA-damaging stress (e.g., ionizing radiation). This would be advantageous in some circumstances and deleterious in others, depending on the relative value placed on cellular survival vs. elimination of potentially neoplastic cells. Thus, this observation raises questions regarding how SIRT1 is itself regulated.

In back-to-back papers in Nature earlier this year, two labs report the discovery that the DBC1 (“deleted in breast cancer”) protein specifically inhibits SIRT1, in turn increasing p53 activity and thereby stimulating p53-mediated apoptosis in response to genotoxic stress. Of the two papers, Zhao et al. have the more elaborate abstract, reproduced below; Kim et al. reach similar conclusions:

Negative regulation of the deacetylase SIRT1 by DBC1

SIRT1 is an NAD-dependent deacetylase critically involved in stress responses, cellular metabolism and, possibly, ageing. The tumour suppressor p53 represents the first non-histone substrate functionally regulated by acetylation and deacetylation; we and others previously found that SIRT1 promotes cell survival by deacetylating p53. These results were further supported by the fact that p53 hyperacetylation and increased radiation-induced apoptosis were observed in Sirt1-deficient mice. Nevertheless, SIRT1-mediated deacetylase function is also implicated in p53-independent pathways under different cellular contexts, and its effects on transcriptional factors such as members of the FOXO family and PGC-1 directly modulate metabolic responses. These studies validate the importance of the deacetylase activity of SIRT1, but how SIRT1 activity is regulated in vivo is not well understood. Here we show that DBC1 (deleted in breast cancer 1) acts as a native inhibitor of SIRT1 in human cells. DBC1-mediated repression of SIRT1 leads to increasing levels of p53 acetylation and upregulation of p53-mediated function. In contrast, depletion of endogenous DBC1 by RNA interference (RNAi) stimulates SIRT1-mediated deacetylation of p53 and inhibits p53-dependent apoptosis. Notably, these effects can be reversed in cells by concomitant knockdown of endogenous SIRT1. Our study demonstrates that DBC1 promotes p53-mediated apoptosis through specific inhibition of SIRT1.

The fact that the DBC1 protein was originally identified as one that is frequently deleted in breast tumors suggests that there are indeed tissues in which unchecked SIRT1 deacetylation of p53 would be a bad thing (i.e., in which it makes sense to kill off damaged cells, even at a cost to regenerative capacity — another example of the evolutionary tradeoffs between regenerative capacity and tumor suppression).

One obvious question is whether DBC1 is also commonly deleted in other epithelial tumors; if so, is there a pattern in the tissue types that develop such tumors? e.g., perhaps DBC1 is particularly important in epithelial populations that, like the breast, lie dormant for much of the lifespan but possess the latent ability to proliferate rapidly in response to hormones — in cells like these, it makes sense to have a relatively “hair-trigger” apoptotic response to potentially carcinogenic insults. In less proliferative tissues, however, the system might be tuned quite differently, with DBC1 levels set relatively low in order to preserve self-renewal capacity even after a manageable level of genotoxic damage.

Next step, of course: What regulates DBC1?

The proteasome is an essential agent in cellular protein turnover, recognizing and targeting polyubiquitinated proteins for degradation — a process that serves both a regenerative function (by turning proteins back into amino acids, the raw materials for proteins yet to be synthesized) and a regulatory one (a protein that has been degraded can no longer act within the cell).

Papa and Rockwell report that reversible proteasome inhibition has long-term effects on the mitochondria, causing diminished energy production and increased generation of reactive oxygen species (ROS) in neurons. This increase in oxidative stress, in turn, results in increased cell death via apoptosis:

Persistent mitochondrial dysfunction and oxidative stress hinder neuronal cell recovery from reversible proteasome inhibition

Oxidative stress, proteasome impairment and mitochondrial dysfunction are implicated as contributors to ageing and neurodegeneration. Using mouse neuronal cells, we showed previously that the reversible proteasome inhibitor, [N-benzyloxycarbonyl-Ile-Glu (O-t-bytul)-Ala-leucinal; (PSI)] induced excessive reactive oxygen species (ROS) that mediated mitochondrial damage and a caspase-independent cell death. Herein, we examined whether this insult persists in neuronal cells recovering from inhibitor removal over time. Recovery from proteasome inhibition showed a time and dose-dependent cell death that was accompanied by ROS overproduction, caspase activation and mitochondrial membrane permeabilization with the subcellular relocalizations of the proapoptotic proteins, Bax, cytochrome c and the apoptosis inducing factor (AIF). Caspase inhibition failed to promote survival indicating that cell death was caspase-independent. Treatments with the antioxidant N-acetyl-cysteine (NAC) were needed to promote survival in cell recovering from mild proteasome inhibition while overexpression of the antiapoptotic protein Bcl-xL together with NAC attenuated cell death during recovery from potent inhibition. Whereas inhibitor removal increased proteasome function, cells recovering from potent proteasome inhibition showed excessive levels of ubiquitinated proteins that required the presence of NAC for their removal. Collectively, these results suggest that the oxidative stress and mitochondrial inhibition induced by proteasome inhibition persists to influence neuronal cell survival when proteasome function is restored.

Here’s what caught my eye about this paper: Mitochondrial ROS production is widely considered to play a causative role in cellular aging. Mitochondria, in turn, accumulate damage over the lifespan, which causes an acceleration in ROS production — a nasty positive-feedback loop. Based on these findings, however, I wonder whether the decline in mitochondrial function could be driven not only from within (by oxidative damage to mitos causing further oxidation), but also from without: We know that proteasome efficiency also declines during aging; it would seem likely that this functional decline could further erode mitochondrial function.

One corollary of this hypothesis is that even if we were able to completely eliminate oxidative damage, we’d still suffer diminution in mitochondrial capacity as a result of proteasomal decline –unless, of course, this decline and its effect on mitochondria operate via a non-oxidative mechanism, in which case eliminating oxidative damage would kill two birds with one stone.

Next Page »

Follow

Get every new post delivered to your Inbox.

Join 44 other followers