Elastin degradation in cardiovascular aging

If you removed every living cell from a human body and looked at the result, you’d still see something recognizably human: bones, of course, and the keratins that make up our skin and hair…and, forming a fine lacework throughout the entire body, the extracellular matrix (ECM). The ECM, which is a particularly prominent feature of connective tissue, consists primarily of large protein complexes that provide structural support (e.g., collagen) as well as elasticity (e.g., the appropriately named elastin).

Elastin is involved in one of the most visible consequences of aging: Over time, elastin is broken down (possibly by proteases secreted by senescent cells), and the skin becomes less resistant to mechanical force. We fight the good fight but eventually gravity wins, and we get wrinkles, wattles, and various other sorts of unmentionable sags. This is specific to later life in part because elastin is only produced during early development and childhood: What you have when you’re an adult is basically all you’ll ever have.

Elastin also has important roles inside the body, most significantly in providing the vasculature and heart with resilience and load-bearing capacity. Indeed, as reported by Pezet et al., mice that are haploinsufficient for elastin display several vascular anomalies and signs of premature cardiac aging. These animals have high blood pressure, narrow and rigid arteries, and cardiac hypertrophy even as young adults. The mice have normal lifespans, but the strain used in these studies all die of a stereotyped set of tumors at an early-to-medium age (for a mouse), so total longevity may be uninformative here.

In light of such findings, it has been suggested (as in this review by Robert et al.) that the age-related breakdown of elastin may place an upper bound on the maximum natural lifespan of the human cardiovascular system (and therefore of any human dependent on such a system).

Solution. “More elastin” sounds obvious, though one would have to be very careful: even though elastin provides elasticity, too much of it might make the arteries and heart overly rigid and unable to perform functionally necessary deformations (think about trying to blow up two nested balloons). Furthermore, excessive deposition of ECM protein in general could result in fibrosis. I would propose a two-fold approach: attack the sources of elastin degradation — calcium deposition, sun damage, and proteases secreted by senescent cells — and in the meantime, figure out how to synthesize more elastin exactly (and only) when it’s needed, so that tissue homeostasis can be preserved without untoward consequences.

 

p53 decline in old age: a cause of late-life cancer?

The p53 protein is one of the principal defenses against tumorigenesis. It integrates upstream signals from the DNA damage response and translates these signals into action: cell cycle arrest, apoptosis, or senescence. The specific outcome of p53 activation depends on a host of variables, including cell type, the extent and nature of DNA damage, and cell-cell signaling.

If p53 were to somehow go AWOL in a cell, it would bode poorly for cancer prevention. Lacking this critical checkpoint control, genetically damaged cells could go on cycling, perhaps developing additional genomic changes that further encourage unrestricted growth, and eventually becoming frankly neoplastic.

A recent study from Arnie Levine’s lab shows that the p53 response to one form of genotoxic stress (ionizing radiation) becomes less efficient in of old mice. If this finding is general to other humans, it could partially explain why the risk of tumors increases exponentially with age . From Feng et al.:

Declining p53 function in the aging process: A possible mechanism for the increased tumor incidence in older populations

Cancer is a disease of aging. The accumulation of mutations in individual cells over a lifetime is thought to be the reason. In this work, we explored an additional hypothesis: could p53 function decline with age, which would contribute to an enhanced mutation frequency and tumorigenesis in the aging process? The efficiency of the p53 response to -irradiation was found to decline significantly in various tissues of aging mice from several inbred strains, including lower p53 transcriptional activity and p53-dependent apoptosis. This decline resulted from a decreased stabilization of the p53 protein after stress. The function of the Ataxia-telangiectasia mutated (ATM) kinase declined significantly with age, which may then be responsible for the decline of the p53 response to radiation. Declining p53 responses to other stresses were also observed in the cultured splenocytes from aging mice. Interestingly, the time of onset of this decreased p53 response correlated with the life span of mice; mice that live longer delay their onset of decreased p53 activity with time. These results suggest an enhanced fixation of mutations in older individuals because of the declining fidelity of p53-mediated apoptosis or senescence in response to stress, and they suggest a plausible explanation for the correlation between tumorigenesis and the aging process.

One important point that might get lost in the shuffle is the observation that the timing of p53 decline is related to the life expectancy of the mice. In an absolute sense, longer-lived mice have high (i.e., normal) p53 activity for a longer time; in a relative sense, however, they have high p53 for the same proportion of the lifespan as short-lived mice.

This gets at a question that biogerontologists like to ponder: Do age-related phenomena occur late in life simply because they take a certain amount of time to get going, or are they coordinated with the overall rate of aging? As with the finding that Aß aggregates form later in longer-lived worms, the p53 mouse finding argues for the latter interpretation (and supports the idea that a 15-year-old dog who happens to get cataracts, arthritis, heart disease, kidney dysfunction and incontinence all at the same time isn’t merely unlucky, but rather a living example of the as-yet-unidentified means by which evolution has synchronized the rate at which the various wheels start coming off the wagon).

Back to the central finding: What might cause such a decline in p53 activity over time? The authors of the paper point out that the signaling kinase ATM, one of the heavy hitters of the most upstream events in the DNA damage response, is also suffering a functional decline with age — but this just shifts the problem up one level. Why is ATM declining?

I wonder whether a contributing factor might be adaptation of the signaling pathways involved. Signaling pathways almost always involve some negative feedback; among other things, this serves to prevent inappropriate activation of a pathway in response to a low baseline level of stimulus, to preserve the dynamic range of the system and reset the threshold so that it can be triggered only by really noteworthy events. (Something like this happens at a circuit level in the olfactory system: ever notice how a really bad smell gets more tolerable over time?)

We know that aged cells accumulate DNA damage (an input into the p53 pathway) and other alterations that might increase their overall stress “tone.” One can imagine that increasing chronic stress might, via negative feedback, increase the thresholds for activation of the DNA damage response at multiple levels, from the upstream events like recognition of damage by ATM all the way down to the principal effector, p53. Consequently, aged cells simply wouldn’t register DNA damage in the same way as young cells: they’ve been hearing that signal too often for too long to get all excited about it. The problem, of course, is that a given amount of DNA damage is just as bad for an old cell as a young cell, and they’d be wrong not to listen.

This is wild speculation, but it makes a series of testable hypotheses, including the potential to turn young cells into old ones by tuning up their chronic genotoxic stress levels and seeing whether they start tuning down p53. It also raises the stimulating counterintuitive possibility that by somehow temporarily turning off the DNA damage signaling altogether, one could cause these thresholds to sink again — in essence, re-teaching an old cell its old tricks.

 

More buzz on bees

For those of you who enjoyed Lev’s post last week about age-related behavioral changes in the eusocial insects, I wanted to point out this dispatch from Behrends et al., who have studied the relationship between cognitive aging and social role in the honeybee worker caste. They find that functional changes in learning performance appear to strongly correlated with social role within the hive, but not chronological age — dovetailing nicely with last week’s news that (in this particular organism) aging and mortality appear to be unconnected to physiological senescence.

 

Bee-havioral Senescence and Anty Aging

At the cellular level, we typically think of senescence as an orderly transition into quiescence or death. Aging, a functional decline in cellular and organismal activity, is often associated with senescence. Both processes are typically influenced by the passage of time, but aging and senescence do not always go hand-in-hand. Two recent papers illustrate the distinction between senescence and aging and touch upon the broader question of the adaptive purpose of these processes.

Rueppell et al. addressed a paradox in the well-studied biology of the honey bee Apis mellifera. These researchers found that, despite a higher mortality risk, chronologically older workers bees do not display significant age-related declines in physiological properties. One explanation might be that older bees drop dead according to a secret internal chronometer that doesn’t influence their function (aging without senescence). Alternatively, the authors suggest that increased mortality risk in older bees may result from age-related division of labor. Worker bees perform various tasks to maintain the colony, including hive-building, nursing of larvae and foraging for food. It is known that foraging bees tend to be older individuals, who may be cannier but also expose themselves to greater risk of predation or mishap. The higher extrinsic mortality of older bees despite an absence of measurable aging might thus result from the dangers of leaving the hive, a privilege accorded only to the oldest of workers.

This hypothesis has found experimental support from a rather cruel study of another eusocial organism, the ant Myrmica scabrinodis. Moron et al. (I kid you not) manipulated the physiological aging of worker ants by injuring them mechanically or with myrmo-toxic carbon dioxide (the bioethics committee must have rejected their application for the use of a magnifying glass). These treatments, while not fatal, shortened the life expectancy of their hapless victims. Injured workers were marked and placed into artificial colonies with equal numbers of unmarked, age-matched healthy workers. The experimenters then carefully observed how the injured ants spent their time and how long they lived. Not only did injured ants die faster than healthy ants, but they spent more of their time foraging on the outskirts of the colony. The authors propose that physiological damage triggers a behavioral change that maximizes the utility of the individual to the colony— individuals that are likely to die soon take on the most dangerous work.

Both studies support the idea that the behavior of eusocial insects undergoes age-related change— this can be thought of as social senescence. In ants, physiological damage that may (or may not) model the natural aging process triggers a behavioral change that makes the best of a bad situation. In bees, cryptic age-related physiological changes promote risk-taking in ostensibly healthy animals. The obvious question is whether these changes can be reversed. For instance, it may be possible to extend the lifespan of forager bees by engineering of colony demographics to decrease the need for foraging.

 

Time flies like an arrow; fruit flies like a young tissue microenvironment for germ line stem cells

Two papers, published back-to-back last week in Cell Stem Cell, address the mechanisms of aging in the Drosophila gonad, specifically germ line stem cells (GSCs). One group studied the testis, the other the ovary, but both come to similar conclusions: Changes in the stem cell niche — i.e., the local tissue microenvironment that surrounds and supports stem cells — are critical to the rate of functional aging.

Yin: In the (ordinally) first of the two papers, Pan et al. demonstrate that BMP and E-cadherin signaling from ovarian niche cells declines with age, and that boosting the level of BMP secreted by niche cells can slow ovarian GSC aging. The authors also show that expression of the antioxidant enzyme SOD, either in niche cells or in stem cells themselves, also slows the rate of functional decline. I find this result less surprising, as oxidation is widely thought to influence aging, though it is worth keeping in mind that stem cell-intrinsic factors are also relevant to aging (for an example in human stem cells, see our earlier post Sucking the life out of marrow).

Yang: Switching gears (and genders) to focus on the testis, Boyle et al. (from the lab of rising star Leanne Jones) uncover a similar story: In the aging testis, niche cells (clustered at one tip of the organ in a region called the “hub”) express lower levels of DE-cadherin and a self-renewal signal called unpaired. As these molecules decline, the number of GSCs decreases, lowering the overall rate of spermatogenesis. As in Pan et al., artificially providing one of the lost molecules (this time, unpaired) rescued the age-related decline in stem cell number.

All fine and good for (unpaired-transgenic) flies, but why should we mammals be interested? Declines in stem cell number and proliferation capacity are thought to be associated with aging and age-related pathology. Stem cell therapy (sometimes envisioned as a simple matter of procuring stem cells of the appropriate lineage and immunological flavor and introducing them into a sick or decrepit body, where they will doubtless know what to do while politely refraining from causing cancer) has been touted as a remedy for disease or even for aging itself.

But here’s a wrinkle, pun not intended: These papers add to the growing body of evidence that stem cells don’t fare well in aged niches. In other words, the stable introduction stem cells into patients will become more difficult as a function of the recipients’ age; since it’s a question of capacity rather than efficiency, the problem can’t be solved by adding more stem cells.

Clinically, this means that in addition to finding immunologically compatible stem cells for each patient and optimizing the conditions for transplantation, we may need to engineer the niche itself in order to cajole aged tissues into accepting a new batch of stem cells — and this will be very hard. We are much better at manipulating cells outside the body and re-introducing them than we are at making genetic changes to particular cells inside a specific tissue architecture while they remain inside the body — and by “much better” I mean that at present we can do the former sometimes and the latter not at all.

Perhaps a better approach would be to make the stem cells less sensitive to the cues they receive from the niche? Yes, that would be nice: immortal cells, freed from requirements for context-specific pro-growth and anti-apoptotic factors, free in the body to do as they pleased…oh, wait.

 

A role for microglial senescence in Alzheimer’s?

Why do amyloid plaques cause Alzheimer’s disease? While it would seem to be self-evident that neurons would prefer not to be surrounded by tangled forest of malfolded, insoluble protein deposits, the mechanism by which these plaques cause neuronal death remains an active subject of inquiry.

Cell-autonomous mechanisms (i.e., those in which the plaques act directly on the neurons that will ultimately die, which are the same cells that produced the Aß and tau protein that make up the plaques) are likely to be most important, but some scholars have begun to consider the cell-non-autonomous possibilities. What if the primary action of amyloid plaques is on another type of cell entirely — such as the ubiquitous, essential, yet still poorly understood neuronal support cell, the microglia? Flanary et al. argue that the presence of amyloid plaques accelerates the process of microglial senescence:

Advanced age and presence of intracerebral amyloid deposits are known to be major risk factors for development of neurodegeneration in Alzheimer’s disease (AD), and both have been associated with microglial activation. However, the specific role of activated microglia in AD pathogenesis remains unresolved. Here we report that microglial cells exhibit significant telomere shortening and reduction of telomerase activity with normal aging in rats, and that in humans there is a tendency toward telomere shortening with presence of dementia. Human brains containing high amyloid loads demonstrate a significantly higher degree of microglial dystrophy than nondemented, amyloid-free control subjects. Collectively, these findings show that microglial cell senescence associated with telomere shortening and normal aging is exacerbated by the presence of amyloid. They suggest that degeneration of microglia is a factor in the pathogenesis of AD.

To summarize: Long-term activation of microglia exposed to amyloid results in telomere shortening (presumably the cells undergo more divisions than when they’re not activated), which ultimately leads to cellular senescence when telomeres become critically short. Consistent with this, senescent cells can be observed in amyloid brains, at higher levels than one would expect as a result of chronological age alone.

The authors do not demonstrate a direct connection to Alzheimer’s pathology, but it’s easy to build a model in which senescent microglia contribute to cell death. Evidence from our lab and others has shown that senescent cells, which accumulate throughout the body as a function of age and genotoxic damage, secrete high levels of dangerous signaling molecules, e.g., inflammatory cytokines, growth factors, and matrix metalloproteases. While the post-mitotic cells in the vicinity of senescent microglia are unlikely to respond to growth factors, the inflammatory factors and protease activity could easily conspire to make life quite unpleasant for the delicate neurons. This would be especially likely if the amyloid plaques are also causing direct damage to these cells.

Several labs are already considering ways to therapeutically eliminate senescent cells, either exploiting the body’s natural methods (immunologically) or using gene therapy. A firm connection between senescence and a scourge like Alzheimer’s (which, unlike aging as such, is already recognized by funding agencies as a pathology) could go a long way toward energizing such efforts.

 

Replication stress vs. lifespan, across evolution

Replication stress, defined as the harmful effects of partially replicated DNA persisting in the nucleus, has a negative impact on the chronological lifespan of yeast, according to a recently published study by Weinberger et al.. The authors propose that a variety of life-extension techniques (including calorie restriction) might benefit cells primarily by preventing them from attempting to synthesize DNA, causing cell-cycle arrest in G1 (when the nuclear DNA is well-ordered and intact) as opposed to S (at which point an arrest results in stalled replication forks and other lesions that will likely be deleterious if they persist).

Chronological lifespan is distinct from replicative lifespan — the former is the time a yeast cell can persist in a nutrient-limited environment, the latter is the number of times a given mother cell can bud. Still, I can’t help but be reminded of recent observations by d’Adda di Fagagna and co-workers that DNA hyper-replication in response to oncogenic stimulation (which also results in stalled replication forks and other structural abnormalities in the DNA) has a baleful influence on the replicative lifespan of mammalian cells in culture. Granted that yeast have no reason to senesce (do they?), the similarities in these two stories suggest a conserved mechanism by which DNA damage (even if self-induced) can limit longevity.

The Weinberger study is at PLoS ONE, by the way, if you’d like to add your own commentary/annotation at the online version of the article.

 

Conference report: SENS3

Okie here, back from the SENS3 conference in Cambridge, and slowly recovering from jet lag.

General thoughts: As a scientist, it is a challenge to present my work to a mixed group of scientists and (particularly well-educated) lay people. Where translational research is concerned, however, I think that lay people do a great job keeping us researchers focused on the prize and not just on (interesting) esoteric points.

As in my previous conference reports (see here and here) I will cover general themes of the meeting, as well as summarizing specific presentations that I found most interesting. Unfortunately I can’t cover them all; what I decide to cover is purely subjective and perhaps even a bit arbitrary. Also, I may skip or gloss over talks/themes that were repeated from the Edmonton conference with little progress.

Themes:

• Biomedical remediation
• Would Healing/Artificial Repair
• Muscle Aging
• Other topics

Biomedical remediation

The fascinating field of biomedical remediation (essentially the brain-child of Aubrey de Grey) is moving along quickly. We heard from two collaborating/competing groups: Pedro Alvarez from Rice and John Schloendorn, a student from Tempe, Arizona being supported directly by the Methuselah Foundation. Pedro is a brilliant environmental chemist/bioremediation guy turning some of his talents on the biological problem of lipofuscin accumulation. The work is progressing rapidly. Both teams have identified strains of bacteria capable of using 7-ketocholesterol (one precursor of the poorly defined lipofuscin) as energy. The next goal is to clone the genes. After that they want to purify the enzyme responsible and feed it to people and see if it will break down our lipofuscin.

My only criticism isn’t with the method, results, or rate of progress (which are all fantastic). My issue is that they are trying to solve a problem that hasn’t been proved to be a problem yet. Lipofuscin accumulation has long been associated with aging in many tissues, but never (as far as I am aware) proved to be responsible for any illness, ailment, or disease. Now, don’t get me wrong, Aubrey makes an excellent argument for this being a serious problem with no traditional biomedical solution in sight, but it’s still just theory. As one of my old mentors used to say, “In this game you’ve got to have data!” Here’s my 2 cents: Now that they’re homing in on the genes, how about cloning the gene and making a transgenic mouse? Might be easier to look at toxicity, long-term affects, and efficacy with a transgenic; though dosage control is problematic with transgenics.

Wound Healing/Artificial Repair

In my opinion, this was the most provocative and promising aspect of the research at SENS 3. Really cool stuff below.

Cato Laurencin is an amazing individual. He is one of those rare clinicians who can aim high-quality research directly at clinical applications. He calls his approach “regenerative engineering.” As I work in a bioengineering department, I sit through a lot of boring biomaterials talks. It was amazing, however, to see someone actually using a few in something practical! In my opinion, this is the reality of regenerative medicine: an innovative surgeon combining technology and knowledge of biology to partially repair injuries such that they will heal as well, or better than they started. Dr. Laurencin showed results from his work on 3D absorbable poly L-lactide (PLLA) scaffolds that seem to promote recovery from surgery much more efficiently than traditional methods. This is a microsphere-based scaffold, which promotes efficient invasion and engraftment of osteoblasts to help repair bone. He is also investigating surfaces with nano-scale grooves, which are more conducive to mesenchymal stem cell proliferation.

Rutledge Ellis-Behnke spoke on his work with SAPNS: Self Assembling Peptide Nanofiber Scaffold. Essentially, he squirts a solution containing these nanofibers into wound sites and reportedly achieves amazing results. He reports dramatic recovery from serious brain injury: both scarless repair of bulk brain tissue removal and reinnervation. In addition, he claims that the nanofibers can dramatically stop bleeding in wounds (he showed video of this). These results are so dramatic that they are almost unbelievable. There are some videos attached to this paper that are pretty darn amazing. The mechanism of action is unknown.

Right along these lines, Robin Franklin gave an interesting talk about myelin repair/regeneration. To summarize the take-home message: the presence of differentiated tissue/cells/debris inhibits efficient re-myelination. If they inhibit clearance of dead myelin by artificial or natural means, re-myelination does not occur. The real trick now is to figure out how to stimulate clearance of damaged myelin (especially in old animals), and the holy grail will be to discover which factor(s) in the damage/differentiated tissue inhibit regeneration.

Muscle aging

Two groups and three speakers addressed the issue of aged muscle, muscle regeneration, and muscle stem cells.

Gillian Buttler-Brown summarized her previous work on human cells, establishing that myoblasts (muscle progenitors) senesce in culture and that cells from old people senesce slightly faster than those from young donors. Interestingly, she showed preliminary work analyzing the “secretome” of myotubes generated from old or young myoblasts. This was inspired by the work of the Campisi lab on the secretome of senescent cells.

Michael Conboy summarized the recent work from the Conboy lab showing how old muscle stem cells can be revitalized after being exposed to a young systemic environment and how embryonic stem cells can have a similar paracrine affect on revitalizing old muscle cells. He then described his recent work on asymmetric cell division in muscle stem cells. Basically, the stem cells tend to divide so that the original copy of the DNA stays with one daughter cell and the newly synthesized DNA segregates with the other daughter cells. This ensures that some stem cells remain behind with original copies of the DNA (which are presumably of higher fidelity).

Another talk from the Conboy lab (by yours truly) was a short study on the telomere regulation of muscle stem cells. Basically, we discovered that truly pure, undifferentiated muscle stem cells (satellite cells) have very high telomerase activity. Furthermore, they continue to fully maintain their telomerase activity and telomeres with age. This supports the idea that muscle stem cells remain intrinsically young, even while their tissue ages around them.

Other topics

If you’re not already familiar with Sangamo, I highly suggest you check out this exciting young company. This isn’t garden-variety gene therapy - it’s gene editing, for lack of a better word. It’s not introducing exogenous DNA into your cells, it’s editing your genomic DNA. Right now (since gene therapy doesn’t work) the best approaches (in my opinion) involve ex vivo manipulation of cells (and the immune system is the most amenable to this approach). As you can imagine, this technology could also be extremely useful for cell culture lab experiments. No more need to create knockout mice just to generate knockout cells. You can do it with many cell types and should work in any species. Right now it is ridiculously expensive to have them generate a cell line for you (I heard $20k a while ago), but they just made a deal with Sigma to start selling the tech to labs, so I figure they are planning on making it large-scale and affordable to researchers. What I would like to hear are ways to apply this tech to make cells better, in addition to curing diseases (like AIDS).

There is an NIA project to test various compounds on the lifespan of mice. No real results are available yet, but if you have a favorite drug, vitamin, or supplement of any kind then you too can recommend that it be tested on mice! Randy Strong of UT-San Antonio gave this presentation.

A great disappointment to me was the cancellation of Rita Effros’ talk. A rather, um, interesting talk was pulled together at the last minute to replace her. A gentleman from a small company collaborating with Geron is selling a “nutraceutical” which is supposedly a potent activator of hTERT expression. For the low, low price of $25k per year, you too can extend your telomeres. They are avoiding FDA regulation by calling it a nutraceutical instead of a drug and by NOT doing any clinical trials. I find it ironic that it’s possible to escape regulation by not doing any testing to ensure its safety. They don’t know what tissues the drug nutraceutical is targeted to. About a dozen clients have been taking the compound for 3-9 months. They report extension of mean telomere length of granulocytes (but not yet other immune cells) and an improvement in vision. There are no placebos or negative controls of any kind (controls would make it an experiment, which would make it a drug). Honestly, I’m really glad that there are people out there willing (desperate enough) to do this sort of self-experimentation and I’m anxious to see the long-term results.

Ruth Itzhaki has made an interesting connection between Alzheimer’s disease and herpes virus infection. According to her results, people with the APOE4 allele and an HSV1 infection (that’s the “kissing disease” with which 90% of people are infected, not genital herpes) were more likely to develop symptoms of Alzheimer’s, and more severe symptoms, than patients with the APOE4 allele alone. She finds that viral load is concentrated in AB plaques and speculates that one HSV1 glycoprotein has a similar structure to the AB protein. Finally, she finds increased phosphorylation of Tau protein after HSV1 infection. Currently, no HSV1 vaccination is approved for use in humans…

Zheng Cui (Winston-Salem, NC) is a man with a mission to cure cancer. You may have heard about his method before: it’s a sort of brute force immunotherapy approach. He isolates white blood cells from a donor and injects them into the “patient” (in most cases a mouse). The granulocytes then attack the tumor and “cure” the cancer. One drawback from this type of therapy is that it requires 10 donors for every recipient. He has done some human work and human granulocytes definitely do the job in vitro.

These are just a sampling a lot of fantastic talks and I wish I had time to write about all of them. The videos of all of the talks will eventually be posted online and I urge you to check them out when they become available at the conference website. There were also a number of talks of questionable scientific quality or virtue. I like to think that the field of aging science is separating from the age-old snake-oil stereotypes, but there was definitely a fair amount of what I would term “pseudo-science.” You can check those talks out too.

On a final note, I would like to make a comment about the state of the art. I would like to see more theoretical and statistical work on which problems of aging are the most pressing/serious ones. I think Aubrey’s “7 deadly things” is a well thought out plan for tackling the problem of universal aging. What I would like to see is some data on which problem(s) are rate limiting. For example, what if solving the problem of “too few cells” (cell death and senescence in aging) would double human lifespan all by itself while all the others put together would barely accomplish the same? The keynote talk (by Ryan Phoenix) included some modeling of how soon SENS treatments could be available, how soon we would need to solve the 7 things in order to treat people alive today, and how often treatments would need to be repeated. This all relied, however, on the assumption than all 7 deadly things were created equally. Everyone agrees that we should take steps to provide the most immediate benefits to humankind, but no one agrees on what these are.

In closing: The humorous poster of my friend and fellow conference attendee George Hinkal, who helped with this piece by encouraging me to add a couple things and helping to clarify a couple others.

 

A hazy shade of Wnt

The Wnt signaling pathway, originally discovered in a developmental context, is now known to play a key role in the homeostasis of many tissues; furthermore, its signaling via ß–catenin (one of Wnt’s several receptors) is perturbed in a variety of tumors. Now two complementary papers have demonstrated that Wnt may play a causative role in aging.

The first study focuses on the klotho mouse model, in which a loss-of-function mutation results in accelerated aging. Liu et al. demonstrate that in wildtype mice, Klotho protein binds Wnt in the serum and thereby antagonizes Wnt interactions with its receptors. In the absence of Klotho (i.e., in the klotho^-/- mutants), Wnt has free reign and augmented activity. The authors further demonstrate that high Wnt signaling accelerates cellular senescence, a phenomenon increasingly implicated in age-related decline in tissue function. (The identification of the specific receptor involved is left as a future exercise.)

The deleterious consequences of a hyperactive Wnt axis are elaborated by Brack et al., who show that increased Wnt signaling (in this case mediated via the less well-studied receptor Frizzled) is associated with a lineage conversion in myogenic progenitors. As Wnt activity increases during aging, muscle cell progenitors switch from a myogenic (i.e., regenerative) mode to a fibrogenic (i.e., inflammatory) mode; this can be prevented with specific blocking antibodies. It is easy to that the resulting increased fibrogenesis, at the cost of regenerative capacity, could cause muscular weakness and sarcopenia in late life.

Happily, some attention is already being paid to pharmaceutical intervention in Wnt pathways, in the context of seeking antagonists that might be useful in the treatment of cancer (see here and here). These efforts currently focus on the consequences of signaling via ß-catenin, which may or may not be relevant to the findings reported above (in the first paper, events downstream of ß-catenin are used as readouts of Wnt activity, but it’s not clear whether ß-catenin is actually required for the increase in senescence; in the second paper, it’s clear that a distinct receptor, Frizzled, at least plays a significant role). Nonetheless, if drug designers aim high enough in the pathway (i.e., at soluble Wnt, where the Klotho interaction is) they might be able to hit two birds with one stone.

 

Delayed aging via increased Arf and p53

A boiling controversy in biogerontology involves whether tumor suppressor genes are beneficial or deleterious with respect to lifespan and aging.

According to the “deleterious” model, tumor suppressors prevent cancer, which is good for survival, but only by eliminating cells (via senescence or apoptosis) required for tissue regeneration late in life — thus, cancer prevention itself bears the seeds of mortality (see our earlier piece, Devil’s bargain: Tradeoffs between stem cell maintenance and tumor suppression, especially the articles about p16 linked at the bottom of the post). Advocates of this position point to cases (e.g., Heidi Scrable’s p44 transgenic) in which elevated p53 axis activity results in accelerated development of aging-associated phenotypes.

In contrast, the “beneficial” camp points out that tumor suppressors detect and suppress damaged cells, whose altered function might have contributed to the aging process if they’d been allowed to persist and proliferate. It is from this perspective that Matheu et al. interpret the data from their transgenic mice carrying extra copies of both p53 and p19^ARF genes.

Delayed ageing through damage protection by the Arf/p53 pathway

The tumour-suppressor pathway formed by the alternative reading frame protein of the Cdkn2a locus (Arf) and by p53 (also called Trp53) plays a central part in the detection and elimination of cellular damage, and this constitutes the basis of its potent cancer protection activity. Similar to cancer, ageing also results from the accumulation of damage and, therefore, we have reasoned that Arf/p53 could have anti-ageing activity by alleviating the load of age-associated damage. Here we show that genetically manipulated mice with increased, but otherwise normally regulated, levels of Arf and p53 present strong cancer resistance and have decreased levels of ageing-associated damage. These observations extend the protective role of Arf/p53 to ageing, revealing a previously unknown anti-ageing mechanism and providing a rationale for the co-evolution of cancer resistance and longevity.

The paper shows that the transgenes cooperatively confer increased cancer resistance: s-p53 (”super”-p53) mice or s-Arf mice have lower tumor incidence than wildtype, and double-transgenic animals have an even lower rate of cancer. Even controlling for the presence of tumors at time of death, the double-transgenic animals show a significantly different lifespan curve: Age at earliest mortality (the time when the first animals in the cohort die) is 300% higher in s-Arf/p53 mice. Furthermore, certain gross functional measurements (tightrope walking, which measures neuromuscular coordination; and hair regrowth, a proxy for regenerative capacity) deteriorate much more rapidly in the wildtype than in the transgenics.

The authors observed that s-Arf/p53 mice generate reactive oxygen species (ROS) at a lower rate than wildtype mice, express higher levels of antioxidant genes, and exhibit lower-steady state levels of oxidative damage in late life. They hypothesize that this slower accumulation of damage protects the animals against the ravages of time and helps them age more gracefully. (Note the subtle difference between this hypothesis, in which the tumor suppressors act to delay damage, and the motivating idea enumerated in the abstract, in which tumor suppressor genes detect and eliminate damage. An alternate explanation of the same data might be that the transgenic animals are more enthusiastically purging damaged cells, leaving behind a surviving population with lower average levels of damage.)

A few qualifications: Maximum lifespan is not increased in the transgenics (in fact, the wildtype mice edge them out, though not significantly). In combination with the delayed earliest mortality, this means that once the s-Arf/p53 mice start dying, they do so faster than wildtype. The authors don’t address this issue, but it does seem relevant to the claim of “delayed aging.”

On another front, the Arf locus is quite complex, and contains at least two genes in addition to Arf (p15^INK4b and p16^INK4a, so the attribution of the phenotype to Arf seems a bit premature. The authors do address this issue, by pointing out that since p16^INK4a is likely to promote aging (again, see the articles linked at the bottom of this post), any delay of aging in their transgenic system is likely to be a consequence of the extra copy of Arf as such. I’m not satisfied by this explanation, since the papers they’re referring to involved knockouts of p16 rather than properly regulated transgenic expression of the gene. Hand-waving? You be the judge.

Finally, a word about the controversy. Do they or don’t they? The authors cite two studies in which higher levels of p53 resulted in accelerated aging and shorter lifespan, and a third that shows rescue of premature aging by elimination of p53. They argue (convincingly, in my opinion) that the apparent paradox between their results and the previous work can be resolved by appealing to a critical difference in the models: In the premature-aging papers, p53 was always turned on at high levels. In contrast, in the slow-aging s-Arf/p53 mice, the genes are present at higher copy number but regulated normally, so the increased dose of tumor suppressor activity is relevant only in the presence of endogenous damage.

(Hat tip to my baymate Francis, for valuable discussion about this article.)