More Details on Bcl-2 Inhibitors as Senolytic Drugs

A few weeks ago researchers announced the discovery of a potential new class of drug capable of some degree of clearance of senescent cells in old tissues. As an approach to treating aging and its associated medical conditions this has long been advocated by the SENS Research Foundation, and is now coming to be known as a senolytic therapy. The paper is published in Nature Medicine, but is unfortunately not open access. The researchers there referred to the most effective drug candidate by its development code name, ABT-263. Another collaborating research group, those involved in identifying dasatinib and quercetin as senolytic drugs in research announced earlier this year, published their own paper on the ABT-263 discovery yesterday in Aging Cell. There they use ABT-263's generic name navitoclax, and this latest paper is open access, so you'll find more of the details laid out and easily accessible.

Cells become senescent as a result of damage, toxins, and stress, ceasing to divide and secreting a range of signals. This is most likely an evolutionary adaptation of processes involved in embryonic development and wound healing that have also come to suppress cancer risk in early old age, shutting down the ability to replicate in those cells most at risk. Senescent cells are destroyed by their own programmed cell death mechanisms or by the immune system, but some evade these fates and linger. The immune system becomes damaged and ineffective itself in later life, and that no doubt doesn't help matters. As senescent cells accumulate in ever greater numbers over the years, the combined effects of their secreted signals become damaging to surrounding tissues, generating inflammation, remodeling important structures, and eventually encouraging the generation of cancer rather than suppressing it. The presence of senescent cells contributes to the progression of near all of the common age-related diseases.

Selective destruction of cells is a major theme in cancer research and other areas of medicine, and destruction of senescent cells looks to be the shortest path to removing this contribution to degenerative aging. A good means of clearing these cells means that we don't have to stop to fully understand how and why cellular senescence causes damage - we can just test removal in the laboratory and look for a beneficial outcome. So far, that is exactly the outcome seen in animal studies. Early this year the first senolytic drug combination of dasatinib and quercetin was tested in mice with mediocre results in terms of percentage of cells cleared and a level of removal that varied widely by tissue type. Nonetheless it produced measurable, significant benefits after just a single treatment. This is a form of narrow, selective rejuvenation, the restoration of some parameters of biology in a living individual to the state they were in earlier life.

The focus of the research and drug screening that identified navitoclax is the inhibition of Bcl-2 and related proteins. These proteins are involved in the regulation of apoptosis, a programmed cell death mechanism. In theory senescent cells should already be predisposed to that fate, so nudging more of them over the line to trigger apoptosis is a plausible approach.

Identification of a Novel Senolytic Agent, Navitoclax, Targeting the Bcl-2 Family of Anti-Apoptotic Factors

Senescent cells contribute to age-related diseases. Much like cancer cells, senescent cells are resistant to apoptosis, potentially protecting them from their own pro-inflammatory secretions, reactive metabolites, and activated DNA damage response. They are instead eliminated by the immune system. We therefore hypothesized that senescent cells depend upon anti-apoptotic defenses similarly to cancer cells. Indeed, our analysis of the transcriptome of senescent human preadipocytes identified pro-survival pathway up-regulation.

Here, we tested if the Bcl-2 family inhibitors, navitoclax and TW-37, are senolytic. Like the combination of dasatinib and quercetin, navitoclax is senolytic in some, but not all types of senescent cells: it reduced viability of senescent human umbilical vein epithelial cells (HUVECs), IMR90 human lung fibroblasts, and murine embryonic fibroblasts (MEFs), but not human primary preadipocytes, consistent with our previous finding that Bcl-xl siRNA is senolytic in HUVECs, but not preadipocytes. In contrast, TW-37 had little senolytic activity. Navitoclax targets Bcl-2, Bcl-xl, and Bcl-w, while TW-37 targets Bcl-2, Bcl-xl, and Mcl-1. The combination of Bcl-2, Bcl-xl, and Bcl-w siRNA's was senolytic in HUVECs and IMR90 cells, while combining Bcl-2, Bcl-xl, and Mcl-1 siRNA's was not. Susceptibility to navitoclax correlated with patterns of Bcl-2 family member proteins in different types of human senescent cells, as has been found in predicting response of cancers to navitoclax. Thus, navitoclax is senolytic and acts in a potentially predictable cell type-restricted manner.

Senolytics could be valuable in treating disorders related to senescent cell accumulation, e.g., atherosclerosis, chronic obstructive lung disease, idiopathic pulmonary fibrosis, osteoarthritis, diabetes, kidney dysfunction, dementias, and neurodegenerative diseases. It appears that the senolytics described so far are limited in the senescent cell types they can target, underscoring the value of testing each cell type involved in particular diseases of interest as part of the senolytic drug development process. We speculate that it may be possible to base selection of senolytic drugs for a particular disease indication on the molecular profiles of the types of senescent cells that underlie that disease. Furthermore, combination treatments for certain indications involving multiple senescent cell types may be optimal in some cases. Overall, our findings support the feasibility of using our hypothesis-driven, bioinformatics-based strategy to develop more, perhaps better senolytic agents. Furthermore, it appears feasible to develop senolytic agents that target senescent cells of a particular type, in a particular tissue, or for a particular indication.


So, this has been niggling at the back of my mind for a while now. If the problem with cancer is that it eventually grows to resist clearance by mutation and selection, wouldn't the same be true in principle for SASP cells?

Posted by: Seth at December 31st, 2015 4:34 AM

We will add this to our candidate list for testing for lifespan. We are already testing Dastinib/Quercetin. Seems reasonably easy to obtain this from our overseas suppliers.

Posted by: Steve H at December 31st, 2015 4:46 AM

The only problem with Navitoclax is the toxicity and side effects. Until a suitable non toxic solution is found it may be better to test a JAK inhibitor to show what would happen if the cells are directly removed. It is not ideal of course as really we want to remove the senescent cells completely.

If Michael is lurking around I wonder what are your thoughts on what we should test to be most useful? We are already testing Dastanib/Quercetin for lifespan and Alk-5 inhibitors.

Posted by: Steve H at December 31st, 2015 8:01 AM

A JAK inhibitor such as the same research group tested here:

It is not of course dealing with the senescent cell SASP just suppressing it but Navitoclax appears to be quite toxic.

Posted by: Steve H at December 31st, 2015 8:16 AM

Reason, great find on both of these studies — thanks very much for these, as for all you do to raise awareness and make the case for intervention in aging in general and the SENS "damage-repair" strategy of rejuvenation biotechnology in particular.

Seth at December 31, 2015 4:34 AM: If the problem with cancer is that it eventually grows to resist clearance by mutation and selection, wouldn't the same be true in principle for SASP cells?

Remember, the essential definition of cancer is that it is a disease of out-of-control cellular replication; it's that rapid replication, along with a high rate of mutation, that leads to cancer's terrible evolvability. By contrast, the essential definition of cellular senescence is that it is a state of arrested cell growth (enforced by particular pathways) in normally-replicative cells. Such cells don't divide, and thus lack any opportunity for natural selection and thus evolution. Additionally, senescent cells don't even have the "normal" rate of mutation undergone during normal cell division in tissue stem and progenitors, let alone the pathological hypermutation present in cancer.

Steve H at December 31, 2015 8:01 AM: The only problem with Navitoclax is the toxicity and side effects.

Well, Sprycel/dasatinib also has substantial toxicity and side-effects (and this page hasn't been updated to include the risk of pulmonary hypertension with dasatinib), but you are intending to test that. Inevitably, any drug based on targeting metabolic pathways instead of stable damage is going to have substantial harmful side-effects through bystander effects on the same pathways' normal function in healthy cells. These drugs have provided good proofs-of-concept for the benefits of ablating senescent cells, but are not true rejuvenation biotechnologies as "damage-repair" agents.

Steve H at December 31, 2015 8:01 AM: Until a suitable non toxic solution is found it may be better to test a JAK inhibitor to show what would happen if the cells are directly removed. It is not ideal of course as really we want to remove the senescent cells completely.

You seem to be suggesting that JAK inhibition will ablate at least some senescent cells; are you aware of any evidence in support of that?

Steve H at December 31, 2015 8:01 AM: If Michael is lurking around I wonder what are your thoughts on what we should test to be most useful?

Well, you already know that I regard Alk-5 inhibition as a poor surrogate for the effects of damage-repair on the systemic environment (in addition to the fact that it's not damage-repair, it only affects a small subset of the changes with age in the systemic environment, and it's then a shot in the dark to as to which specific beneficial effects in rejuvenating the systemic environment will be obtained "for free" from the repair of which particular kinds of cellular and molecular aging damage, and will be confounded by the side-effects of the drug itself). I would certainly say that adding any already-validated (if crude) senescent-cell-clearing agent (or new cocktail thereof, if carefully designed) is of greater value than that. You want to see the effects of damage-repair, not to test a pharmacological mimetic of a subset of the downstream effects of such repair.

Posted by: Michael at December 31st, 2015 12:06 PM

This probably isn't the place or article to really ask this, but does the Hayflick limit end up applying to any of this stuff here? Like eventually could you only clear senescent cells so many times before it wouldn't matter anymore? Or is it not as important in the overall big picture of aging? I read different things about whether or not it's important.

Posted by: Ham at December 31st, 2015 12:06 PM

Ham, I don't understand what you're getting at here, but the following may either answer it somehow or help you to clarify your question. First, remember that what happens when cells in culture reach the Hayflick limit is exactly that they enter into cellular senescence. Second, the Hayflick limit is essentially an artifact of cell culture: in vivo, cells are exposed to much lower levels of oxygen (≈3-5% instead of 20% (atmospheric) O2), and fibroblasts and most other cell types (including most tissue stem and progenitor cells) almost never replicate 50 times absent some secondary replicative stimulus.

The senescent cells you see in vivo are instead the result of activation of proto-oncogenes, replication driven by inflammation or local tissue damage, disease-related oxidative stress, obesity, etc, sometimes (particularly in the case of HSC) accompanied by relatively high levels of default replication over the lifespan.

At some point, of course, in a dramatically-extended lifespan, you really would start running down the replicative capacity of stem and progenitor cells in particular, which is yet another beneficial effect of the seeding of tissue stem cell pools already required for the only plausible strategy for the permanent prevention and cure for cancer on the table as yet.

Posted by: Michael at December 31st, 2015 12:41 PM


Yeah, my question was strange. It really wasn't clear at all (I've been super tired lately haha), but you've been helpful as usual. So I was reading this article (granted, it's not the most scientific or technical, but I saw the thing from Hayflick):

“There are many people who recover from cancer, stroke, or cardiovascular disease. But they continue to age, because aging is separate from their disease,” Hayflick said. He added that even if those causes of death were eliminated, life expectancy “would still not go much beyond 92 years.”

"In the 1960s, Hayflick found that human cells divide only 40 to 60 times, after which they stop — even when their division is paused and then allowed to resume. This discovery, called the Hayflick Limit, indicates that even if a drug like metformin were effective at suspending aging, the end game stays the same.

I know what I'm trying to ask, but having a hard time putting it in words, but essentially it was in regards to the last part, and how important was the hayflick limit as far as senescent cell removal goes, and aging in general.

Posted by: Ham at December 31st, 2015 2:13 PM

Ham, back in the 1960s when the Hayflick limit was discovered, nobody knew what stem cells, let alone pluripotent stem cells, even were.

As Michael points out, the things that cause old cells to stop dividing aren't the Hayflick limit, and when they do stop dividing, we can (theoretically, hopefully, sometime damn soon) replace them in all cases with fresh, pluripotent cells that can.

Posted by: Slicer at December 31st, 2015 4:35 PM

@Reason: yeah, I'd seen that — but that study shows that JAK inhibition inhibits SASP, whereas when Steve said that they were thinking of testing "a JAK inhibitor to show what would happen if the cells are directly removed," I misunderstood him to be saying that he thought that such an agent would actually directly remove them, whereas I see now that he means rather that it would mimic (some of) the effects of doing so by tamping down the SASP.

@Ham: Yeah, Hayflick the man has some very strange and frankly illogical views. In addition to regarding "aging itself" as distinct from the diseases of aging (a formulation that has some theoretical utility IMO but breaks down in practice the more granularly you try to draw the distinction in an actual aging body), Hayflick often seems to regard "aging itself" as distinct from both the cellular and molecular damage of aging (despite frequently drawing the analogy to rust and invoking the Second Law of Thermodynamics) and the metabolic effects of aging, leaving one to wonder where exactly he imagines this ghostly "aging" thing to be located in the aging biological machine. Moreover, he steadfastly refuses to grant that aging can be biomedically modified in any way whatsoever (even going so far as to argue that Calorie restriction, for instance, does not retard the rate of aging); in that context, he therefore is even less inclined to accept the Ship of Theseus analogy as to why the repair and replacement of cellular and molecular aging lesions by rejuvenation biotechnology would lead to the arrest and reversal of degenerative aging. And, he claims to believe that successful intervention in aging and the ensuing increase in human lifespan would be undesirable, which (flowing downward from sour grapes) I suspect lies upstream of his arguments for its impossibility.

He had been saying these kinds of things already for decades when Hayflick and I (in two separate but interrelated articles) and then Hayflick and colleagues had this out several years ago; he has retread the same poor arguments in print several times subsequently, and also in a panel discussion with Dr. de Grey and others at the Gerontological Society of America conference in 2009; and he is still saying them now.

Posted by: Michael at December 31st, 2015 4:47 PM

Thanks Slicer. And yes, the sooner we can do that, the better.

Posted by: Ham at December 31st, 2015 4:52 PM


Ham, my 2 cents, I believe, as Leonard Hayflick suggests himself, there can be several 'types' of aging. PAthological types of aging (the one where diseases are prevalent) are often due to do inflammation cause and gene mutations/gene network instability. All the while, intrinsic 'basal' regular aging the same time. They are tied together but can also be uncoupled.
The problem with stem cell renewal is that they are bound by replicative senescence base pair loss and at some point, telomerase can't help them, stem cells lose telomeres, same goes for fibroblast who at a certain point can't divide anymore (having reach replicative senescence).
There are 3 main pathways : Replicative-senescence causing telomeric DNA damage/loss, Oncogene-induced damaged causing telomeric DNA damage/loss and 'random DNA' damage causing telomric DNA damage/loss.
Telomere loss is causal to intrinsic aging; stem cell included in that; except immortal stem cells and immortal cancerous cells who highjack telomerase or use other cunning tactics so become immortal. p53, p16 is lost and immortality is feasible (Hayflick talks about this replicative finality where cells can't overcome p53/p16 replicative senescence/growth arrest (nearly complete telomere loss and demethylation/5-methylcytosine loss), otherwise they transform to tumors, who themselves evade death by maintaining their telomeres around the short 2 kb telomere size from constant illegal mutated telomerase/hTERT/catalytic subunit use.
So yes, Ham, I believe, that is one big problem we have not solved and Hayflick is right, overcoming replicative senescence is going to be hard; for now it is all about health extension. My take to increase Hayflick limit, and live way behond anything possible, is lying somewhere in the Redox control of Telomeres, Telomeric DNA repeat, Nucleotides and Chromosomes. Cells that evade Hayflick limit, are immortal, and do not accumulate lipofuscin. When SENS removes lipofuscin and AGEs, and protein carbamylation, adducts and the whole plethora of damages; then we will possibly be able to surpass Hayflick replicaive end problem.
Again this is just a opinion. : D

Posted by: CANanonymity at December 31st, 2015 5:21 PM

@Steve H - why don't you contact Oisin Biotechnology and ask them if you can test their method of senescent cell removal out? If this technology is a trade secret, then they'll probably say no. But if they have a patient pending on it, then they'd just be getting some free extra research when you guys carry out and in mouse test.

It's much easy said than done, but why not use AAV and crispr to replace the p16 gene in each mice's muscles with then gene for the prodrug Kirkland's group used plus the p16 gene? Basically an in vivo somatic gene therapy of the germline gene therapy they carried out (except on in muscles, or whatever cell type are targeted by the AAV vectors you use)?

Posted by: jim at December 31st, 2015 6:27 PM

@Michael Yes this is exactly what I meant about JAK Reason explained it right. Its a mimic not actual removal but shows in a crude way what might happen.

I suspect we will test Navitoclax to be honest just thinking about the next batch of testing and the options.

Alk-5 wise your comments are noted.

It might be better to test a second Senolytic if we cannot find a workable solution.

@Jim we would be happy to test this for them, we are a 501 foundation and as you say they would be getting free research.

Posted by: Steve H at January 1st, 2016 2:40 PM

We have a new promo film now as we are gearing up towards fundraising. We are entering the Mouse Prize with Methuselah hopefully too.

Posted by: Steve H at January 5th, 2016 2:02 AM

I am a concerned VA patient with only the practical experience of having serious
type 2 diabetes,bilateral stenosis of inner & outer carotid arteries, one side
of which was "rootered out" in a carotid endothelial procedure in May of 2013.
That same year, I was given a groin access angiogram, then told by the VA/UCLA
vascular surgical team that they were urging me to go along with their plan to
perform maximum complexity open heart, open rib-cage, surgery with a first year
risk of more than 10% mortality. They implied that my chances of living another
5 years without that surgery were only about 50%, & likely would be weakened
by the invasive nature of the plan. I had consulted an outside medical opinion
& examination of the data disk suggested that I had well developed collateral
arterioles ( more than 50 ) that accounted for my ability to climb stairs in
5 story buildings & hustle to airline departures, & catch up with cabs. At the
start of 2015 I weighed 220 lbs. & had a glucose level of 7.9 HbA1C.
I needed to bring my 18 year long challenge of T2 Diabetes under control. I found 3 other vets who had diabetes and all of us were on metformin. I was on
the extended release version, which allowed me to take, not 500mg daily but
1500mg without G.I. distress. As a former pre-med student, I had some ability
to follow the discourse on line about the positive "health-span" effect of that
respected ( aside from the G.I. issues) prescription med. So then I resolved to
try a synergistic approach by taking Ptero-pure, the USDA developed, synthetic
version of pterostilbene (which had been found years before in blueberries, &
in a south India tree bark). Within a few days of adding the ptero-pure to the
mix, my daily fasting finger-stick numbers began to plummet. From the 7.9 A1C
& finger-stick values of 150 or so fasting, I was soon down to about 100 fasting, then the nineties, then mid eighties, finally coming down to a current
A1C of 5.1. My weight spiralled down in tandem, with the A1C values. From the
220 5 months ago I have easily been able to control my food intake to an average of 1450 to 1600 daily calories, while avoiding most bread, rice, & starches. All of the others have lost meaningful weight. I have lost 41 lbs. in this modified CR fasting schedule.
& went from a waist of 40in. to below 33 inches. Visceral, and subcutaneous fat has mostly gone away. and my 74 year old skin have mostly tightened up.
The VA/UCLA vascular surgeons are going to work on the left side of my carotid
stenosis in April this year. My official VA medical records document the results in terms of blood chemistry, A1C, & weight. After 18 years I am also
not obese at 179 on my 6' 1.5" frame. Our group is also now testing Niagen &
recently added quercetin, just to see if any good effects can be observed, since we are former fighting men, not mere lab rats. We hope that the VA will
eventually see the value of having human volunteers, trying synergistic substances, along with promising pharma-meds like extended release metformin. Acting before conventional double-blind,
randomized clinic trials, as a kind of slightly risky scientific RECON OPERATION!
We are talking to key Congressional medical issue tasked staffers, in Washington, who
are serving on the VA medical sub-committee...Never forget...there are nearly 26,000,000 U.S. veterans, many of us are good observers, & went on with our
civilian education & careers.

Posted by: Allan Silliphant,VA medical research activist at February 1st, 2016 1:56 AM

Yeah, this is an old thread, but maybe someone gets prompted when new posts are made.
Say an effective senolytic was developed tomorrow.
Would there be an anticipated risk of a tumor lysis type of syndrome from dying senescent cells?
How does the burden of these senescent cells increase percentage wise as humans age; is this even known?

Posted by: Nyles7 at February 12th, 2016 2:17 AM
Comment Submission

Post a comment; thoughtful, considered opinions are valued. New comments can be edited for a few minutes following submission. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.