Methylation Aging Clock: An Update

Methylation of DNA is the best-known mode of epigenetic regulation (turning genes on and off).  Methylation patterns are stable unless they are actively changed, and can persist over decades, even across generations.  

Four years ago, biostatistician Steve Horvath of UCLA identified a set of 353 methylation sites that are best-correlated with human (chronological) age.  These are sites where genes are turned on and off at particular stages of life.  A computer analysis of a gene sample (from blood or skin or even urine) can determine a person’s age within about two years.

Two reasons the Horvath Clock is important.  First, it is the best measure we have of a person’s biological age, so it provides an objective measure of whether our anti-aging interventions are working.  Say you’re excited about a new drug and you want to know whether it really makes people younger.  Before the Horvath clock, you had to give it to thousands of people and wait a long time to see if fewer of them were dying, compared to people who did not get the drug.  The Horvath clock is a huge shortcut.  You can give the drug to just a few people and measure their Horvath (methylation) age before and after.  With just a few dozen people over a two-year period, you can get a very good idea whether your drug is working.

Second, there is evidence and theory to support the idea that the methylation sites that Horvath identified are not just markers of aging but causes of aging.  That means that if we can figure out how to get inside the cell nucleus and re-configure the methylation patterns on the chromosomes, we should be able to address a root cause of aging. (Before we get too excited: “Gene therapy” has been around 20 years but is still in a developmental stage; “epigenetic therapy” is what we need, and it does not yet exist, but is technically feasible using genetically engineered viruses and CRISPR.)

The write-up below is taken directly from two talks that Horvath gave, 2016 at NIH in Maryland and just last month in Los Angeles.


In 2012-2013, three papers appeared proposing the idea that the deep cause of aging (in humans and many other higher animals) is an epigenetic program [Johnson, Mitteldorf, Rando].  Genes are turned on and off at various stages of life, producing growth, development and aging in seamless sequence.  (A fourth paper by Blagosklonny proposed a similar idea, but focused on the role of a single transcription factor controlling gene expression (mTOR) and shied away from the conclusion that natural selection might have preferred aging affirmatively.  Here’s an earlier presentiment by Blagosklonny.)

It’s a powerful hypothesis that proposes to resolve evolutionary and metabolic questions alike.  It contains a seed of a prescription for anti-aging research—although epigenetics has proved to be so complicated that practical modification of the body’s gene expression schedule may require a lot more groundwork.

Unbeknownst to any of us working on these theoretical papers, Steve Horvath was already working on calibration and measurement of the epigenetic aging clock, and he published his basic result by the end of 2013.

One remarkable property of the Horvath clock is that it is more accurate than chronological age for predicting who will contract aging diseases and who will die.  Even though the clock was derived with an algorithm that matched the output clock age as closely as possible to chronological age, the result proved to contain more information than chronological age.  “In deriving the clock, chronological age was used as a proxy for biological age.”  People whose “methylation age” is greater than their chronological age are likely to suffer health deterioration and to die sooner than people whose methylation age is less than their chronological age.  

Horvath has openly shared his methodology and his computer program.  Based on the Horvath clock, a California company began last year to offer a commercial test for methylation age.  You can send a blood or urine sample to Zymo Research.

 

Candidate aging clocks

Horvath describes how he came up with the idea of a methylation clock by a process of elimination, beginning with four candidate clocks:

  1. Telomere length
  2. Gene expression profile
  3. Proteomic data
  4. DNA Methylation

In detail:

  1. Telomere length – This had been measured easily and cheaply for more than a decade, but its correlation with chronological age (and with mortality) is not strong enough to be useful as a biological clock.

  2. Gene expression profile: Which genes are being transcribed into RNA at a given time?  This can be measured by extracting RNA, and turns out to be highly tissue-specific.  In other words, it varies according to which part of the body you’re looking at.
  3. Proteomic data:  Genes, once transcribed, are translated into proteins.  Some of these proteins stay in the cell while others circulate through the body.  Gene CHIP technology measures levels of different proteins reliably and inexpensively.
  4. DNA Methylation: Easier to measure than (2) or (3). Methylation is only one of many mechanisms controlling gene expression, but it is one of the most persistent.  Horvath found that a subset of DNA methylation sites seems to be characteristic of age no matter where in the body they are measured.

What is DNA methylation?

Adjacent to many genes is a promoter site, a location on the same chromosome which stores temporary information about whether the gene is turned on or off.  Promoter sites contain the base sequence C-G-C-G-C-G-C repeated.  This is called a CpG island (where the “p” just tells you that the C is linked to G on the same strand, rather than being linked across strands, in which C is paired with G.)

C stands for “Cytosine”, and the Cytosine molecule can be modified by adding an extra methyl group (CH3) to form 5-methyl Cytosine.

The cell has molecular workers that are deployed to go around specifically adding methyl groups in some parts of the DNA or removing them in others.  The bottom line is that methylated Cytosine is a sign that says “don’t transcribe the adjacent gene.”  When the methyl groups are removed, it is a signal that the gene are to be transcribed once more.

Enzymes called methyl transferases are deployed to precise regions of the genome to turn genes on and off.  Methylation can be transient.  There is evidence for circadian cycles of methylation.  Or it can be quite long-lasting.  Methylation patterns can persist for decades, and are copied when cells replicate, so that methylation patterns can be passed to offspring as part of one’s epigenetic legacy.  Inherited methylation sites are the exception however; most of the genome is programmed fresh with age-zero, pluripotent methylation patterns when egg and sperm cells are generated.

 

How the methylation clock works

Using a standard statistical algorithm, Horvath identified 353 CpG sites that were most strongly correlated with chronological age, no matter where in the body he looked.  The same algorithm provided 353 numbers to be multiplied by methylation levels at each site, then added up to produce a number.  The number is not directly a measure of age, but in the last step a table is used (an empirically-derived curve) to associate the number with an age.

This is the raw output of the function before it is transformed into an age.  Notice that methylation changes very rapidly during the first 5 years of life, gradually slowing during the growth phase and straightening out to constant slope after about age 18.

Even though the Horvath clock was designed to be independent of what part of the body DNA was drawn from, some variations appear.  Most noticeable is female breast tissue, which ages faster than the rest of the body, and brain tissue, which ages more slowly.  Blood and bone tissue tend to age a little faster.  (Sperm and egg cells are “age zero” no matter the age of the person from whom the germ cells were drawn.  Placentas from women of all ages are age zero.) Similarly, induced stem cells (using the 4 Yamanaka factors) have zero age.  In contrast, a similar treatment can change one differentiated cell type into another, for example, turning a skin cell into a neuron.  This does not affect epigentic age.

Liver cells tend to be older than the rest of the body in people who are overweight, and younger than the rest of the body in people who are underweight.  Other tissues don’t seem to show this relationship.  For example, fat cells do not have older methylation ages in people who are obese.  And, perhaps surprisingly, weight loss does not reverse the accelerated methylation age of the liver (at least, not within the 9-month time frame of the one study looking at this).


Studies have been done correlating methylation age with various diseases and, of course, mortality.  Corrections are made for every kind of environmental factor, including smoking, obesity, exercise, workplace hazards, etc, called collectively the “extrinsic factors”.  The result is that methylation age rises with extrinsic factors, and independently methylation age is also correlated with intrinsic (genetic) factors that affect lifespan.  Horvath estimates that genetics controls 40% of the variation in methylation age (as it differs from chronological age).

Men are slightly older than women in methylation age.  This is already evident by age 2. Delayed menopause is associated with lower epigenetic age. Cognitive function correlates inversely with methylation age of the brain.

Speaking before Horvath at the same conference, Jim Watson claims there are many supplements and medications that can slow the Horvath clock.  The one he focuses on is metformin, which, he says, has epigenetic effects via an entirely different pathway from lowering blood sugar (the purpose for which it has been prescribed to tens of millions of diabetics).

 

Here’s a curious clue:  There is a tiny number of children who never develop or grow, and continue to look like babies through age 20 and perhaps beyond.  These children have normal methylation age.  Whatever it is that blocks their growth, it is not the methylation changes in their DNA.  Does this mean that there are other epigenetic controls, more powerful than methylation, that control growth and development?  Or does it mean that children with this syndrome have normal epigenetic development, but something downstream from gene expression is blocking their growth?  Conversely, Hutchinson-Gilford progeria is caused by a defect in the LMNA gene which causes children to age and die before they even grow up.  Hutchinson-Gilford children have normal methylation ages by the Horvath clock.

Radiation, like smoking and exposure to environmental oxidation, tends to age the body faster.  This is independent of methylation age—which is unaffected by radiation.  Neither smoking nor radiation exposure affect epigenetic age.  HIV also accelerates aging, and HIV does affect methylation age.

Methylation age and telomere age are both correlated with chronological age, and they both predict mortality and morbidity independent of chronological age.  But the two measures are not correlated with each other.  In other words, the information contained in the methylation clock and in measures of telomere length complement one another to offer a better predictor of future aging decline than either of them separately.

Diet has a weak effect on methylation age.  Very high carbohydrate, very low protein diets are noticeably terrible.  Beyond this, there seem to be two sweet spots: one for the Ornish-style protein-restricted diet and one for the Zone/Atkins style diet.  Weak evidence to be sure, but suggestive that they both work.

“The epigenetic clock is broken in cancer tissue.” [ref]

 

Building on the original clock

The original clock was optimized to track chronological age, and yet it fortuitously provided more information than chronological age.  In a second iteration, Horvath set out explicitly to track biological age.  He used historic blood samples from the 1990s, and paired them with hospital records and death certificates to search for methylation sites that correlate best with aging-related health outcomes.  The result was the phenotypic clock, DNAm phenoAge.  This uses 513 methylation sites to predict

  • all-cause mortality
  • cardiovascular mortality
  • lung disease
  • cancer
  • diabetes
  • (loss of) physical strength
  • (loss of) cognitive ability

On the drawing board:  An epigenetic clock specialized to work well with skin and blood cells, (which are the most accessible).  (Enough skin cells can be scraped painlessly from the inside of your mouth (buccal epithelial cells) to do a DNAm test.)

 

Connection to Parabiosis and Plasma Transfusions

Several groups have begun to experiment with transfusions of blood plasma from a young donor as a possible path to rejuvenation.  Horvath reports an encouraging finding:  Sometimes older people contract a form of leukemia that requires a blood and marrow transfusion (including the stem cells that give rise to new blood) from a donor.  The finding is that after this treatment, the blood of the patient continues to show the methylation age of the donor, not the patient.  

 

Epigenetic Aging and Telomere Aging Bound to a See-Saw Relationship

(This was the most exciting new result for me personally, because it relates to an idea I have held dear for more than a decade.)

Methylation age is older or younger than chronological age in different people, generally by about +2 years.  40% of the variation is due to genetics.  Some common genetic variants can make the clock run faster or slower.  The most prominent genetic variants link telomere aging to methylation aging.  The faster your epigenetic clock runs, the longer your telomeres.  The slower your epigenetic clock runs, the shorter your telomeres. [preprint]

There’s a word for this in the genetic theory of aging.  It’s called Antagonistic Pleiotropy.  Back in 1957, George Williams theorized that the genes causing aging ought to have simultaneous beneficial and detrimental effects.  That would explain why natural selection has permitted aging to occur, despite the fact that it cuts off fitness.  Williams said: Nature had no choice but to accept the genes that cause aging because there was no other way to get the benefits of these same genes (which he surmised ought to enhance fertility).

My theory of Antagonistic Pleiotropy is that it is not a situation of “forced choice”; rather, aging is important for the health of the community, and mother nature has been faced with the dilemma: how to keep aging in place despite efficient natural selection against it on the individual level.  Aging is so important to the community that evolution has been motivated to find ways to keep it in place, despite the short-term temptation for natural selection to favor those with longer lives (thus greater opportunities to leave offspring).  In my hypothesis, evolution invented pleiotropy to address this problem. The telomerase-epigenetic clock connection is an example.  There is no physically necessary connection between telomerase and epigenetic aging, but the two have evolved a see-saw link so that it is more difficult to mutate aging away.

This also relates to my coverage last fall of the telomerase-cancer connection.  At the time, I was scratching my head, why should genetic variants that lengthen telomeres be associated with higher rates of some cancers?  Here is a clue: The same genetic variants that lengthen telomeres also accelerate the epigenetic aging program.  The specific example of a cancer that is most closely tied to higher telomerase levels is melanoma, which is a cancer that is less sensitive to age than other cancers.  People tend to get melanoma earlier in life than other skin cancers. Therefore, I predict that other pleiotropic links will be found between these genetic variants that promote longer telomeres and other mechanisms linked specifically to melanoma.


The Bottom Line

All these data in a field so new is a tribute to Horvath’s industriousness and to the promise and fruitfulness of a new methodology.

The data so far suggest that methylation programming is a big part of the driver of aging, but not the whole story.  Smoking affects life expectancy, but it doesn’t affect methylation age.  Weight loss benefits life expectancy, but it is invisible to methylation age.  Most curious are those children who fail to develop, or age prematurely, even though their methylation age is progressing on schedule.

What does it mean that radiation ages the body without advancing the methylation clock?  Perhaps that accumulation of damage is part of the phenotype of aging, though I remain hopeful that the body remains capable of undoing that damage even late in life, if it is re-programmed to want to do so.  What does it mean that AIDS advances the aging clock?  Perhaps that the immune system is a central signaling mechanism in the aging process.

So, it’s “methylation plus”.  Plus what?  Not just methylation plus damage”; though we can certainly shorten our lifespan with radiation or smoking, we can’t increase our lifespan by avoiding toxins.  “Methylation plus other epigenetic programs”—this would be my first guess.  “Methylation plus mitochondrial state” would be a close second. Methylation is all in the nucleus, and the cytoplasm of the cell seems to store independent information, and can even re-program the state of the nucleus, as suggested by parabiosis experiments. There is also evidence for“Methylation plus telomere shortening”. 

219 thoughts on “Methylation Aging Clock: An Update

  1. I always enjoy reading your articles. One thing from this one that confused me: “Diet has a weak effect on methylation age. Very high carbohydrate, very low protein diets are noticeably terrible. Beyond this, there seem to be two sweet spots: one for the Ornish-style protein-restricted diet and one for the Zone/Atkins style diet. Weak evidence to be sure, but suggestive that they both work.”

    The Ornish-styple protein-restricted diet IS a very high carb low protein diet, with seems contradictory here.

    • You have it contradictory. The only diet that prolongs life is a high-fat diet consisting of saturated, hydrogenated fats. Have you heard of the Mediterranean diet and the French paradox?

        • There is a lot of non-science based opinion when it comes to diet advice, often passed on like established fact. If you simply look at cohorts with the longest healthy lifespans and highest number of centenarians, you find they eat mostly plant based whole food diets with very little meat and very low saturated fat. Longest lived = California Adventists, much has been published about their diet. See: https://www.crsociety.org/topic/11303-okinawans-vs-adventists-is-it-the-low-calories-or-vegetarian-diet/
          But I’m with Josh on the idea that diet alone is not the answer, real life extension progress will come from emerging technologies and cutting edge science.

          • Yes, the influence of fat in food is small. Hydrogenated / non-hydrogenated fats can affect life expectancy by 10-15%. This is negligible. But if the life span were extended by 10%, it would have far-reaching economic consequences. Expenditure on pensions and healthcare would increase and living standards would drop by 30%. Waste generation and environmental pollution would increase.

        • dobrotad I thought you were joking at first. Please provide some references on why you think saturated fat extends life. You should get to know the work of Dr. Michael Greger, author of “How Not To Die” and founder of the non-profit site nutritionfacts.org. Unrefined plant-based whole food diets are recommended by most doctors and clinicians with any clue about lifestyle and nutrition. (My first degree was in Sports Medicine and nutrition). The balance of evidence, the longest lived communities (as Gordo referenced) as well as the recommendations of leading nutrition experts Dr. Michael Greger, Dr. Dean Ornish, Dr. Nathan Pritikin, Dr. Caldwell Esselstyn, Dr. Joel Fuhrman, Dr. Garth Davis, Dr. Neil Barnard, Dr. Joel Kahn all lie with a plant-based diet eliminating or minimizing animal fats. High levels of dietary saturated fat over time is associated with cardiovascular disease, stroke, dementia, sexual dysfunction, and most recently diabetes. There is no paradox its all in the data, as is often explained by Dr. Greger and his team who painstaking go through every study in every (English language) peer-reviewed nutrition journal in the world each year.

          • You may be interested in the PURE study published in the Lancet, a very well done , prospective study over a 10 year period covering 18 countries , and it showed an increased overall mortality from high cacbohydrate diets while virtually all fats showed a lower total mortality rate.

            Association of fat and carbohydrate intake with cardiovascular disease and mortality in 18 countries… Dehghan et al. Vol 390. 2050-62. Lancet. 2017

          • Paul, there is a big difference between a high carb plant based whole food diet (best health/highest longevity) and a high carb processed food diet (refined sugar and flour based -> worst health). Fiber and phytonutrients are likely key here, as well as total caloric intake and BMI. So any study of “normal” populations with regard to macronutrients is going to be fatally flawed from the start. That’s why I prefer cohort studies and tracking biomarkers of health.

          • Clearly many studies tend to paint all carbohydrates with the same broad brush and this is a real problem. So you make a valid point. But aren’t you a bit surprised that all fats, including saturated ones, actually lowered the total mortality rates. This should result in a total overhaul of the AHA diet recommendations, but of course it won’t.

    • I don’t believe the Ornish diet is very low protein and Josh did specify very low protein which you would not think would be a big deal but perhaps it is?

  2. In the text book for my genetics course (Human Genetics, Ricki Lewis) we learned that DNA replication has extremely high fidelity, generally producing less than 1 error in a billion. In contrast, replication of methylation sites can be as much as 1 in 40, and the resulting “epigenetic drift” may also be a primary driver of the dysfunctional aging phenotype.

    I head read several others proposing finding a way to use CRISPR/gene therapy to ‘enter the nucleus’ of every cell and both repair the DNA and reset its methylation state to solve this problem.

    But isn’t there a much simpler solution here? Consider the following:
    1) Most of our non-neuronal cells are constantly being replaced and regenerated over time by stem cells and various progenitor cells.
    2) The Yamanaka factors can effectively reset methylation state to age 0 (as well as telomere length as I understand it).
    3) Gene therapy still runs in the 100s of thousands of dollars, while stem cells can be harvested or induced for thousands of dollars.

    Why not stem cell therapy using the Yamanaka factors to cleanup methylation state? Clinicians could even repair DNA via CRISPR in the test tube, then coax the cells back into the type of stem & progenitor cells required and re-introduce them into the body.

    Imagine that unlike gene therapy, which becomes less effective over time as cells turnover, this therapy would become MORE effective over time as cells originating from those stem cells proliferated. A person’s average methylation age might be getting younger again with time. Heck, while you have the cells out of the body, might as well repair the mtDNA and you have a single therapy that addresses 5 of the biggest culprits in aging – DNA damage, methylation state/epigenetic drift, telomere length, mitochondrial dysfunction, and stem cell depletion.

    • … and if you know anyone with plans to do something like this, let me know. I am ready to become a principal or an employee in that company. And a client, of course 😉

    • If the error in replicating methylation in daughter cells is really as bad as that, then you could see how tissues could become terminally differentiated over time, and stem cells lose their potency.

      You’d also need to show this can happen in non-dividing cells too, possibly though DNA repair?

      • I am guessing what Horvath is really measuring is correlated to some sort of work done to repair DNA.

        That would explain how the stem cells of young donors can grant older recipients blood and bone marrow with a ‘younger’ Horvath age. It also explains how breast tissue ages faster according to Horvath, i.e. hormones stimulate growth, and how telomerase does the same in immortalized cells. Radiation totally smashes up the DNA, irreversibly arresting the cell, so that stops the whole process and explains how this doesn’t affect the Horvath clock.

        • Oh and losing weight cannot magically make liver cells do negative work (they’ve already done extra DNA repair because of past abuse).

    • Very good suggestions PhoenixQ. Only point 2 can we reset using Yamanaka factors safely in vivo?

      Josh your trademark post on very interesting subject. Brief yet every line was insightful.

      Horvath has not received the accolades that he deserves. Discovered a new way to meter aging and continues to research in a very intriguing space. Am a big fan and look forward to his next revelations.

  3. BTW I had my methylation state sequenced by Epimorphy at the consumer-facing site https://www.mydnage.com/. I think Epimorphy split off from Zymo to focus on this consumer testing service.

    The results are very interesting but my goal as you suggest was to have a baseline with which to measure effectiveness of anti-aging interventions.

    • These were my results from Orsiris Green. 56 at time of test. Had been on Rapamycin for 3 months. Can’t make heads or tails from it. It said my weighted age was 54.8.

      TOM1L1 Gene:
      Target Of Myb 1-Like Protein 1
      This gene is found on human chromosome 17, and produces a protein that appears to be important in cell signaling. The associated gene is found in a high concentration in humans in pancreas and thyroid gland, as well as in cells of the lungs.
      Using the information from only this gene, your age estimate would be 79.06 years.
      NPTX2 Gene:
      Neuronal Pentraxin 2
      This gene is found on human chromosome 7, and produces a neuronal petraxin protein. These proteins are found at neuronal synapses (in the nervous system) and are similar to C-reactive proteins, which are major components of the body’s innate immune system. The protein produced by NPTX2 plays a role in the formation of excitatory synapses. The epigenetic changes that occur with age in this gene appear to be different in men and women, which is why we take gender into account when estimating your age.
      Using the information from only this gene, your age estimate would be 47.51 years.
      EDARADD Gene:
      Ectodysplasia A Receptor Associated Death Domain
      This gene is found on human chromosome 1, and produces a protein that plays an important role in embryonic development. The protein itself is essential for proper interactions between two cell layers in a developing embryo (called the ectoderm and the mesoderm), which form the basis for many of the body’s organs and tissues. Interactions between these cell layers are essential for the proper development of several bodily structures, including skin, hair, nails, teeth, and sweat glands.
      Using the information from only this gene, your age estimate would be 29.22 years.Note:
      The difference among the age estimates from the three genes listed above averaged 33.2 years. This number is larger than the normal average difference of approximately 9 years. (In other words, the individual age estimates were not very consistent.) While this result is anomalous, it does not necessarily have any broader implications beyond its effect on our calculations. Your overall age estimate, listed near the top of this page, is a weighted average of all three age estimated, calculated according to our mathematical model.

  4. I can’t seem to grasp methylation age yet. Is more active methylation a good thing as one ages and is the slowing down of methylation correlated to aging faster? Are certain micro nutrients critical for methylation to occur? What are the defining characteristics of aging methylation?
    Been struggling with this for a while. Thanks

    • Methylation is only a result of an aging organism. As oxidized-damaged mitochondria work at a lower rate, and as a result of age-accumulated cadmium blocking enzymes containing Zn and Mg, the gene will adapt to these degraded conditions. The cadmium concentration in the brain is lower than in the body and therefore the concentration of cadmium in the brain is still rising and causing a decline in mental capabilities. Cadmium is a cumulative poison. I can not imagine how it would be possible to move cadmium from a place with a lower concentration to a higher concentration (kidney).

        • In Central Europe, a study was conducted on humans in a cadmium-contaminated area. There is increased abortion, maternal mortality, fertility, and reduced maximum life expectancy. With current drugs, it is not possible to wash out cadmium from the brain.
          If you would find a two-step cure for removing cadmium from the brain, you should at the same time have a “rejuvenation method” that can prolong life for decades.

  5. I can’t give a reference, but I accessed it with Google browser – the report is that mefformin alters the DNA methylation – genome wide!1 the report should be easy to find; but of course that still doesn’t get to the detailed effects – but we do know metformin has been shown to sometime extend lifespans!!

      • I just listened to Jim Watson’s talk. The above article is exactly what he referenced as evidence of Metformin’s effect on slowing down epigenetic clock.

        • Problem is all these interventions can only ever SLOW down the rate of epigenetic aging, not reverse it.

          I’ve never heard of anything, short of reprogramming cells that can turn the clock BACK.

          • Heterochronic parabiosis can do what you suggest. I completely agree with you, Mark, in that stretching out a lifespan (which is sort of what we’re talking about with most forms of life extension) also stretches out senescence. But it seems that cellular reprogramming does reverse cellular age, and evidence shows (remember that partial evocation of Yamanaka factors in live animals that apparently set back aging) that the aging ‘marks’ (if such exist) are superficial compared to the ‘marks’ (epigenetic) that control the differentiated state of the cell. If somehow those methylation sites are the ’cause’ of aging (which I doubt) then de-methylating them should result in a reversal of aging.

          • I’d agree that the evidence in your paper and elsewhere, showing old organs become young in young recipients and young organs in old recipients become old, would suggest some form of reprogramming is occurring in vivo. Seems like organs in transplant patients should be Horvath’s next target.

  6. Dear Josh,

    You make an assertion in your article which may or may not be true. In the history of attempts to explain aging, a host of factors which change with age have been attributed to be the ’causes’ of aging. So telomeres, which shrink with age have been called the cause of aging, and experiments in which ectopically expressed telomerase in fibroblasts prevents their senescence and seemingly gives them immortality. (Quite in distinction to your observation that telomere length is directly related to aging rate, and certainly contrary to efforts to increase telomere length in order to extend lifespan.) So mitochondria become ‘leaky’ with age, producing ROS and also become inefficient, producing less ATP per carbon taken in than the mitochondria of younger cells. Both of these age-related changes have been said to ’cause’ aging. Similarly, hormones such as klotho which decline with age are also said to be causes of aging (I don’t know why no one is taking klotho to assess its effects?). Now, Josh, you say that the increase in methylation at certain positions of the genome may be responsible for aging (other increases, such as the increase in senescent cells showing senescence associated secretory activity (SASP) have also been suspected to cause aging, so it’s not only a decrease with age that might cause aging.) So here is my question, why hasn’t anyone tried to prove that the methylation of these Horvath sites causes aging? While in whole animals, we cannot effectively target all cell nuclei and change their methylation patterns even with a CRISPR-Cas9 enzyme modified to remove DNA methyl groups, such an experiment could be done using stem cells. As you say, there are several characteristics that change with aging including mitochondrial degradation (measurable) and transcription profiles (measurable), even telomere length (If telomere length has a relationship, inverse or not, to DNA-methylation-assessed age, one might expect that removal of methyl groups would change (?) telomere length.) In any case, applying CRISPR-Cas9 to stem cells in vitro is very doable – and assessing the changes in the ‘age’ of the cells is also pretty straight-forward (for example, changes in the rate at which stem cells cycle, as well as those other changes mentioned), so why hasn’t it been done?

    • I think it’s a very worthwhile idea for an experiment, Harold. I know a lot less about experimental biochem than you do, so you’ll have to tell me what is feasible.

      If I understand you correctly, the core idea is to
      * Grow stem cells in culture
      * Modify their methylation patterns using CRISPR, consistent with young Horvath age.
      * See if the cells cycle more frequently

      • Basically correct. But better, obtain stem cells from older individuals rather than just growing them in culture – it’s at the level of the body that aging occurs. And monitor mitochondria and transcription – that can be done on the level of individual cells. I wonder though why no one is trying klotho?

        • Hi Harold.

          How would you suppose to supplement klotho? i have done some studying and its implications are big, not only for health, but also its correlation with intelligens.

          As many might know this is not my usual source, but just as an appetizer see wikipedia’s first couple of sentences:

          “Klotho is an enzyme that in humans is encoded by the KL gene.[5]

          This gene encodes a type-I membrane protein that is related to β-glucuronidases. Reduced production of this protein has been observed in patients with chronic renal failure (CRF), and this may be one of the factors underlying the degenerative processes (e.g., arteriosclerosis, osteoporosis, and skin atrophy) seen in CRF. Also, mutations within this protein have been associated with ageing, bone loss and alcohol consumption.[6][7] Transgenic mice that overexpress Klotho live longer than wild-type mice.[8]

          It is my experience that trying to alter enzymes is also harder then one might think… any ideas is welcomed

    • Another experiment that comes to my mind as I read your proposal: Swap the nucleus of an old cell into a young cell. Swap the nucleus of a young cell into an old cell.

      Does the cytoplasm reprogram the nucleus, so the age of the resultant cell is determined by the cytoplasm? Or does the nucleus reprogram the cytoplasm (and mitochondria) to determine the phenotopye of the cell?

    • You’d have to control for telomere length as cells with long telomeres cycle faster than cells with short telomeres. Cells immortalized by telomerase cycle fastest of all. So you’d have to separate out telomere effects from methylation ones. In vitro experiments with immortalized cells showed the Horvath clock continued to tick, but without any slowing to the cycling of examined cells. But these were somatic not stem cells.

  7. Maybe methylation isn’t what it is made out to be vis-a-vis aging. Transcription can’t start without the promoter being exposed to the polymerase complex and this requires unmasking of the site, histones being a key player (Sinclair and others)

    • Agreed that methylation of DNA is but one part of the genetic code – and where it ranks in the hierarchy (is it an initial step, or does the histone code set it up?) isn’t known – the best thing about it is it’s easily assessed by the bisulfite method so easy to measure.

    • Why, in the course of evolution, should “rise the clock” in the controlling rate of aging?
      What is the evolutionary advantage of organisms that can increase or decrease the aging rate?
      None. There is already a programmed cell death. No gene that accelerates aging, nor any gene that adds age does exist. When you add the missing hormones (androgens, progesterone, GH) or missing neurotransmitters, you do not “rejuvenate”. You only work more intensely as if you were young.

  8. The problem with Horvath’s analysis I think that it only looks at methylation at known or predicted promoter sites (Illumina27k set). So I think this is essentially a development clock not an aging clock. And because aging is a continuation of development it incidentally predicts aging and other problems associated with a more aggressive development problem.
    An aging clock IMHO should look at all the methylation sites in the whole DNA and not only at promoters. It should look at for example centromeric regions where there are a lot of transposable repeat elements. Once unmethylated by aging they can be transcriptionally active and cause genetic instability. I dont know if these regions have been looked at and no result was found or simply there isnt a cheap method to collect the sufficient amount of data on whole genome methylation.

    • I don’t know but it seems to me if you do whole genome sequencing with and w/o bisulfite treatment you should be able to discover all methylation sites – it’s no longer that difficult – but Horvath found that most of the many thousands of sites were not ‘informative’ of aging. If aging is an extension of development – personally I always think of aging as post-adult development – then why isn’t it also an aging clock?

      • Hi Harold,

        I think Horvath used Illumina27k data because in 2012 datasets were available only with this chip. Illumina450k was released in 2011, so I doubt he had large datasets with it. He simply transformed the avilable Illumina450k datasets into Illumina27k and did his work on the reduced dataset. But maybe I am wrong he should answer this question.
        I think it is not an aging clock because it does not capture effects like the generic demethylation and unmasking of transposable elements. These might be driven by different mechanisms than the organismal development.

          • Hi Mark, Gabor,

            Jim Watson and Vince Giuliano over at their blog seem to suggest that the Horvath clock does take into account all changes in methylation:

            I quote:

            “This global hypomethylation feature of aging was used to make up 160 of the 353 CpG sites in Steve Horvath’s “DNA methylation clock” that were found to correlate closely with chronological age, using a computer algorithm with an elastic net regression model. (See Section B above.) Interestingly, the hypomethylated sites chosen by computer algorithm included an over-representation of hypomethylated cytosines in GpG shores (the borders of CpG islands), not gene bodies, transposable elements, or satellite repeats”

            The title of the article is ‘Aging, health and disease – view from the DNA Methylome’ at their anti-agingfirewalls’ blog. Really worth going through, particularly section B where they review 4 other epigenetic clocks. They also establish comparisons between gene mutation in accelerated ageing diseases and the silencing of those same genes as we age. Very informative

            I haven’t gone through Horvath’s paper recently so I cannot confirm what the case is exactly. However, I would agree with Harold that if ageing is the continuation of development then saying that these clocks measure ageing or development is saying the same thing. There could be sites missing, but those found so far would still be informative.

          • Hi Adrian,

            that Watson page is really good, really sums up my own understanding but much more throrough than I have ever been.
            The way I think about methylation is that it is determined by the histone modifications. There are much more histone modifiication enzymes than DNA methylation ezymes.
            And at the end of the day the histone code is determined by DNA motifs I believe. So with careful analysis of the whole genome we may discover the blueprint for aging in the noncoding part of the DNA.

    • Yes the original Horvath clock is based on methylation levels of 353 CpG sites on the Illumina 27 k array. But now he has a new phenotype clock based on methylation levels of 513 CpG sites. This clock can predict not only all cause mortality, but also mortality from specific cases as well morbidity based on his talk last month. But commercially I think only 353 CpG sites get looked at.

  9. Hi. Thank you for an exciting post!
    The epigenetic age of sperm was shown by Dr Horvath to be on average approximately 10 years younger than the man, but not zero in his paper (figures 3 i and j). I have not seen measurements of the epigenetic age of human oocytes.

  10. Fascinating! Thank you Josh. In particular the insight on the possible causative role of the methylation sites in aging. At a last Sept. meeting in Basel (Switzerland), the folks of InSilico Medicine presented a comparison of the chronological age prediction accuracy of their DDN deep learning algorithm compared with Horvath’s, as well as others, and got R2=0.82 with MAE (error) of 5.55 years vs. Horvath’s r=0.96, R2=0.93 and MAE=2.7 years which probably is not that bad considering the practicality of using their DDN algorithm and the ease to follow the relative changes during proactive interventions. More in their paper: “Deep biomarkers of human aging: Application of deep neural networks to biomarker development”.

  11. Great article Josh! I am a 73 yr. old biochemist w/ a high A1C so have been on Metformin for a few months and some berberine for a bit longer. I’ve read that berberine (supplement vs. script like Met.) is a Metformin mimetic so I’ll be watching to see how close of a mimetic berberine is, if anyone happens to look into this.

  12. Placentas age, rapidly, in 9 months.
    Sperm and eggs must age, since there are deleterious effects of advanced maternal and paternal ages.

  13. My way of thinking is, the more upstream that we can go, the better. And I am anticipating the discovery of something we don’t know about yet, farther upstream than anything we currently know about, via big data and artificial intelligence.

    Genetic variants associated with longevity and/or slower epigenetic clock? That’s not the path to finding it. Centenarians look their age. Centenarians die.

    Forget the caloric restriction and anything related to it, too. A true anti-aging treatment would make us be able to eat like we did when younger, without the consequences that we undergo when older.

    I tend to get frustrated with all the talk of biological pathways where development and aging are intertwined. I am especially intrigued by James P. Watson, M.D. saying that HERVs loose their silencing as we age. Something in the genome that goes back millions of years in human evolution? And maybe, just maybe, no pleiotropy? It’s just bad, all bad, when those HERVs get turned on? That’s my favorite theory.

    At any rate, I love Horvath’s epigenetic clock. I think that it will help with the big data and artificial intelligence that I am talking about.

    P.S. I made one mistake, and tried to post as NYLA rather than NY2LA. Now, I am unable to post using NY2LA. So I am starting over with a new user name (my initials) and different email address.

    • I too have lots of problems posting and have been forced to use a new email address. But even that does not always work. I agree with the upstream idea. Any cause of aging can’t have anything upstream of it. Therefore telomeres can be a cause of aging, because shortening them only requires cell replication. Losing a youthful methylation pattern could possibly be a cause of aging too, if DNA repair or copying involves errors in maintaining methylation. Everything else always seems to have another cause.

    • As Mark W, I also experimented many issues to post. Just in case I solved some simply avoiding posting web addresses (copy/pasted or even simply written down as text).

  14. I think that the post was very interesting and informative, and may turn out to be quite useful, but in my mind nothing is more accurate or predictive than just asking someone how they feel, and if they feel any better on any given regimen. For instance, I have exercised, followed a strict diet, and taken numerous supplements for years, and while I feel pretty good with all of those measures, I’ve never felt younger. That changed with rapamycin. After only 6 months I would say that I feel about 12 years younger and am once again able to do things that I could do then, so there is no test that would convince me otherwise and so what’s the point really.
    I would be willing to bet that , if asked, people could estimate their “ real age “ with amazing accuracy, and even how long they are likely to live. Likewise, they are able to tell if any given regimen is making them younger, or not.

    • Agree. Like Bill Andrews always says, just look at someone and you pretty much know their age. Even celebrities with face fillers and plastic surgery can’t really fool the discerning eye.

      • I’ve heard people here with the opinion that they’re conquering aging, but when you look at their photos, you don’t see that. There are probably things that can’t be ‘rejuvenated’, for example cataracts (I think), but replacement isn’t hard (and in my own case was an improvement over the eyesight Nature gave me) but things like skin texture, muscular strength, hearing, etc. can be rejuvenated as shown in numerous studies.

        • You’re right. Nothing could be rejuvenated. When you look at the list of recommended anti-aging products, it is obvious that they do not do too much finery.
          I use the cheapest lifeextension products for EUR 400 a month for 20 years. You can see to me that I look 10 years younger. Those supplements I can swallow for 70 more years before I die. Perhaps antioxidants that do not prolong life by 100% but 200% will be invented over time.

          • I have age-related macular degeneration so I can not deal with stupidity but I have to take really strong anti-aging drugs.

          • Just on the faiths worse then death, i have cared for and seen a lot of those. Many years in a emergency/ and after care of people with traumatic brain injury, i am talking about sometimes sadly heartbreaking faiths.. but also incrediable life encouraging in many ways!

            take this as it is, anadoctal at best, ive worked with a lot of different patients, i prefer to call them, “residents”/ “beboer” in my native tounge. These are patiens that needs care 24/7. well the point is some of these people also have diabetes, and ive witnessed it really having a positive overall effect, treating these people with metformin, despite these people being very brain damaged also, the metformin seem to help “heal” the brain too, Incredible stuff i believe.. we will see much more in the future…

    • Has anyone else on Rapamycin seen their max heart rate go up? I’ve noticed about a 3-4% increase in MHR since I started Rapamycin in Jan 2017. As far as I can tell MHR should go down as you get older.

      • That’s a very interesting observation and something that I’ve never checked. As you probably know, the maximal heart rate is approximately 220 minus your age. So if you’re say 50 then your maximal would be about 170 bpm. So you’re saying that now it’s about 177 bpm, which would now make you the equivalent of a 43 year old Not too shabby! I’m going to now check mine. I’ve felt for a while now that rapamycin has a beneficial cardiac effect and that that accounts for the improved stamina and endurance that I’ve experienced. Maximal heart rate does absolutely decline with age so your observation is a very astute one.

        • Does maximal heart rate decrease with fitness training? Or maybe it depends how it’s defined. The better shape you’re in, the lower your heart rate for the same intensity of exercise. But the better shape you’re in, the higher the intensity you can tolerate.

          • Not to my knowledge. With fitness the resting heart rate drops due to an improved stroke volume. The equation is this : cardiac output = heart rate x stroke volume, so when the stroke volume improves the rate is able to drop without any loss of cardiac output. But the maximal heart rate uniformly decreases with aging and is the most important factor for the loss of stamina and fitness and is believed to be the reason for the feebleness leading to nursing home admissions.
            The formula of 220 minus the age is a fairly accurate determinate of maximal heart rate, and is therefore a marker that could easily be followed and monitored with regards to anti-aging protocols.
            I have suspected all along that rapamycin improved my cardiac output, but I thought that it was from an improved stroke volume, but now I’m really wondering if an increase in maximal heart rate is the answer. That would be a pretty exciting discovery by LO. I’m going to have to find a decent heart monitor and check this out.

      • I’ve seen 190-191 on some of my peak efforts on a bike. Very difficult to reach unless you are chasing someone or being chased. Before dosing low 180’s would be the highest I would see. Low 180’s more common now. 57 years old so that’s not too bad. I was just wondering if anyone else is seeing something similar.

  15. I do recall a nuclear exchange or something similar (perhaps it was cancer?) where the cytoplasm had the major effect. I’ll see if I can find it.

    • Interesting. There is a related recent paper by same author.

      “Nucleolar expansion and elevated protein translation in premature aging”

      Nucleolar size increase during normal aging and this is exacerbated in HGPS cells. The authors conjecture that it might be due to the general heterochromatin loss which also occur during aging.

      Nucleolar expansion leads to increase of rRNA production and ribosome biogenesis which, in turn, increases protein synthesis. One interesting take is that mTOR inhibition could be a way to alievate the problem.

      “Since mTOR regulates both ribosome biogenesis and translation initiation, inhibition of the mTOR pathway could alleviate phenotypes of HGPS23 by pushing back against the influence of progerin on ribosome biogenesis and translation”

      Unfortunately, the authors tempered that prolonged treatment with rapamycin might have significant side effects and do not seem to be aware that intermittent (weekly) usage can address them.

      • This has allways puzzled me, how does people making these studies (with a huge knowledge of cell biology,) not know about intermittent rapamycin? it seems so weird to me… maybe these people are just really into what they are doing that they don’t look at the bigger picture? mmh.

        • Most people, including and especially most scientists, won’t even consider rapamycin due to its “risks”. Here are the options:
          A. Rapamycin = low risk
          B. Aging = Certain Death

          They opt for B.

  16. A logical next step from the Horvath clock would be to look at the methylation sites chosen by the statistical algorithm and ask, “Which proteins are being promoted, and which are are being suppressed?”

    The methodology is purely statistical, but the logic behind it is that methylation controls gene expression. So we are curious to know what genes are controlled by these methylation sites?

    During the Q&A of the January 22 video, Horvath answers that he has looked for transcription correlates and has not found any! This is a very interesting null finding. It means that there’s something basic we’re missing or misunderstanding about the effect of methylation. Horvath mentions the possibility that methylation affects transcription not just at adjacent genes but at distant sites as well, in ways that we have yet to appreciate.

    • My idea about this is that the Horvath clock being a development clock affects stem and progenitor cells mainly. So one would have to look at the expression profiles of stem cells and then it may be transient. Because these sites belong to bivalent chromatin, they only get transcribed when needed.
      The other part of methylation changes, those that the Horvath clock does not capture, for example generic demethylation over age, they would surely pop up in the expression profile data.

      I would rather be interested in histone modification patterns and the DNA motifs that make the histone modifying enzymes attach to the specific differentially methylated sites.

      • Gabor – you’ve misread the clock’s essential methodology. The original 353-site methylation clock was explicitly designed to work with any tissue in the body, not stem cells. The new 513-site PhenoAge clock is designed to work with white blood cells, not stem cells.

        • Hi Josh, I know that. Thats not my point. The methylation changes are uniform in each cell, but they only take any effect in stem cells and even then only transiently in case of injury, division, growth stimulus, etc. But I guess they only check the expression profiles in PBC, when they made the claim that there is no correlation between resting expression profile and the methylation clock.
          What makes cells function less well in old age is probably the generic demethylation, which is distinct from the Horvath clock. Otherwise Horvath could have made the clock only out of the hypermethylated sites only, since there is no information content in the demethylated sites – they get stochastically demethylated over the whole DNA.

          • I am very much in agreement with GaborB. We know the Horvath changes have no detrimental effect on somatic cells, at least based on in vitro work. And we have lots of hints the body is stimulating stem cells with increasing futility as we age. So I suspect we’ll find epigenetic changes associated with the Horvath clock affect cell differentiation. This could be linked to why demethylation appears so common across tissue types, but methylation is more tissue specific. We should be able to prove this with in vitro work.

          • Random errors in methylation has replaced random DNA damage as the ’cause’ of aging – but aging is still seen as an accumulation of errors; an aging cell is a damaged cell. If that were the case, rejuvenation would be impossible, rejuvenation as shown in heterochronic transplantation, or parabiosis (although cellular reprogramming might also remove methylated sites) – so I don’t believe that that sort of “damage” causes aging. Although I might be accused of heresy, I don’t believe that DNA is responsible for aging (at least directly), I believe that though it may act as ‘clock’ – a clock does not control time, it only records it.

    • Mark and Gabor –
      You paint a picture in which differentiated cells are footsoldiers which perform a function locally, but which have no wider influence. Maybe the master pituitary glands and neuroendocrine glands in the brain tell the footsoldiers what to do without backtalk.
      My guess is that there is some of this, but there is also feedback. Humble skin cells and muscle cells are sending out signals into the blood. I learned last week about exosomes–packets of such chemical signals encapsulated in lipids.
      If we’re really lucky, there’s a central clock that keeps track of age and the rest of the body just follows orders. But I think it will turn out that a part of the story is the distributed clock, the democratic clock, the consensus clock that is coded in the epigenetic state of a trillion cells.

      • Absolute agreement Josh. The work of Cai (at Einstein in the Bronx) showed that changes in the hypothalamic clock (SCN) were a response to distress signals from cells far removed from the brain. It is know that mitochondrial distress signals are transmitted through the bloodstream as is the DNA damage response – by such factors as the HMG proteins. The assumption of most is that aging occurs at the cellular level and spreads upwards through the hierarchy to affect tissues, organs, etc., but I think it’s a ‘two way street’.

        • Hi Josh, Harold. I don’t think we’re as far apart as you think. I’m sure that factors in the blood are responsible for de or re differentiation of cells as required at various sites in the body. But maybe epigentic changes in cells make them less responsive to these signals as we age, hence the rise in systemic inflammation?

    • The body is composed of cells and nothing else. The DNA in the cells does not change when we go from 30 yrs to 40 yrs to 60 yrs. so what could be the cause of aging, it is the gradual lack of access to certain favorable genes and gradual access to unfavorable genes. moreover the cell function largely does not change when we go from 30 yrs to 40 yrs, whereas dna methylation changes. therefore aging at least from 30 yrs to 40 yrs is decoupled from function. But the real question is how does it happen gradually

      • I have never studied mitochondria. Can someone tell me? When dividing cells, will the old mitochondria be divided into two parts, or will there be completely new mitochondria according to “cellular DNA” in the newly created cell?

  17. Josh am perplexed by the way the blog has changed over the past year or so.
    The blog is called “Playing the Game for a longer life”. And shows an image of playing chess.

    My take on this is that the emphasis is on what we can individually do to lengthen productive healthy lifespan.

    But in recent posts the emphasis has been on big science and the various companies that are doing scientific research to stop programmed aging. A lot of posts & comments are now full of scientific gobblegook well beyond my understanding.
    And so we are just potential ‘consumers’ of science based companies with new processes to sell…

    It’s not a change I think for the good.

  18. Bill- You make a good point. I’ve noticed it too. The blog follows my interests, which is the reason it’s gotten more technical. By why is it becoming more about the biology of aging and less about the practical business of optimizing health? First, I’d say it’s because there are so many blogs out there that do a good job with personal health and fitness. Second, it’s because what we know about how to optimize health doesn’t change very fast. It’s only once or twice a year that we learn something new that tweaks our understanding of how to care for ourselves. So I lean more toward reporting the science, which excites me and is changing fast enough to keep me busy.

    I really appreciate the feedback. One thing I can promise to do is to report news that really affects our personal habits with high priority. And another is that I can listen to your suggestions for topics you’d like to hear more about.

    – Josh

    • This is a great blog. As yes, we all love the the direct and immediate intervention possibilities most. But sometimes we have to delve into the science to get those insights.

    • Thanks for the reply. I will stay in the loop and if a significant aspect comes to mind, I will lob it our way…

      It’s true that there are other health & fitness blogs out there – like Rogue Health for example….Or Dr Mercola’s blog.. But many have built biases and assumptions about what is best..and contradictions abound…Which can be pretty confusing..

      Whereas here you are always clear about your biases while being open to hearing evidence that contradicts them…

      That’s why i would regret this blog disappearing like Alice, down the rabbit holes that lead merely to scientific ‘wonderlands’

        • The last time I saw directions of a fragmentary map, twas somewhere ‘over the rainbow’. But not in Oz. That’s where I live and I’ve looked hard enough here already.
          Ummmm !
          🙂

          ‘Low carb’ seems the way to a longer healthy life at least as promoted on some other blog..

          But how to reconcile that with an individualistic inate tendency towards eating less meat ? How do you put those together Josh ?

  19. Aslan, your remarks to Dobrota are crude, cruel and not funny. I suspect Dobrota’s English isn’t good – why would you go out of your way to insult her and make fun of her?

    • Aslan did not mind bad English. Google Translate translates it wrong. The contents of the comments made him sick. He was annoyed by the fact that I am one of the few people in the world who stayed 10 years younger after using antiaging medicine.

  20. Another class of epigenetic modulator is microRNA. What is most interesting is that they found plant microRNA can go through GI and enter sera and organs.
    L. Zhang, et. al., “Exogenous plant MIR168a specifically targets mammalian LDLRAP1: evidence of cross-kingdom regulation by microRNA,” Cell Research, doi:10.1038/cr.2011.158, 2011.
    https://www.nature.com/articles/cr2011158
    MicroRNAs from the food we eat can directly target our gene expression!!!

    • The usual meme is that rice is bad for you because it’s starchy and initiates an insulin spike, which leads indirectly to faster aging and downstream health consequences. But this article says that there are micro RNAs in rice that are absorbed directly into our bloodstream with direct and detrimental effect on our lipid metabolism.

      What a surprise! How could such a connection evolve?

      • One experiment they can do is to isolate the circulating microRNAs of the young blood and inject them into the old to see if it extends life. They could also inject the circulating microRNAs from the naked mole rats into the regular rats to see if it extends life.

  21. There are a few arguments in favor of epigenetic clocks as measuring ageing. As opposed to only measuring development (they can measure that too):

    1. Horvath himself looked into this, and all DNAm clocks correlate with frailty and can predict age expectancy at a given point. With his latest Phenotypic clock being the most accurate, according to his talk in the video Josh linked above. This is the strongest argument.

    3. The 353 sites clock is conserved in Chimpanzees but not Gorillas. This shows that this ageing/development clock is unlikely to be stochastic in any way. It is probably a developmental program controlling the whole lifespan of a species.

    4. The conventional view of development in humans is that it stops in your late teens or at most in your 20s. So if this clock continues to tick past this age, decade after decade and the only common visible change is ageing, how would these clocks not measure ageing? They must do, at least, in some form.

    So one interpretation here is that development past your youth is what we commonly call ageing. Of course if you don’t favor theories of programmed ageing you will be skeptical of this. Ultimately, whether it is a more ‘directed’ program or a pseudo-program may not matter much if it just happens to behave like one.

    • As I said previously, and full disclosure, I am an advocate of programmed aging; aging is post-adult development. (So I think.) Now the question is whether these methylation changes are determinative, or merely reflect the state of the developmental program. The state of methylation of those sites, I’d guess (especially as they seem to make no difference re. transcription) are the hands of the clock – what about the mechanism that moves those hands?

      • Well Horvath claims his clock is associated with mortality, so it’s at least correlated with other age related changes. But we know from both his and others’ work that telomerase immortalised somatic cells continue to age according to his clock, but are not impaired or damaged by this in any way. So my and GaborB’s deduction from this was that it must affect stem cells by contributing to their loss of stemness. This would then cause rising inflammation and other serious problems that could kill you. But we could be wrong. It might be that long telomeres are stabilising the chromatin so that other downstream changes from this methylation are prevented. Or it might be that the Horvath changes are downstream of something else that is the real cause of aging. But then if that’s true, why the null effect on immortalised cells in vitro? Its a mystery thst can probably only be solved by a much more complete analysis of epigenetic changes in the genome. I would say at this point that it looks like a programmed change to me, though you could argue it’s just ‘drift’.

        • Why would random drift have the same effect on all cells’ methylation state such that you could predict their age from looking at specific sites? So the question is do these immortalized cells age? If age represents the increased probability of death (by one definition) then they don’t age. If age is marked by increasing mitochondrial ‘leakiness’ and inefficiency, then they don’t age. If these immortalized cells display the same transcriptional profiles, generation after generation, then they don’t age, (don’t know if they do or don’t, but I suspect they don’t otherwise the cells would not be immortal). So then maybe these methylated sites simply mark the ‘passage of time’ (though perhaps something other than strict chronological time, some intracellular time (perhaps measured by the energy expended in metabolism), that just happens to correlate with aging changes in intact animals? We know aging happens in intact animals, and that two attributes correlated with aging, Horvath site methylation and telomere attrition are present in cell grown in vitro? Perhaps those other attributes of aging are not present in vitro, and the body directs their appearance in vivo? There seem to be multiple pathways in organismic aging, but maybe only two are present in cells grown in vitro? Does anyone have a reference to those Horvath papers that show that immortalized cells still age? Sorry about all those question marks – I wish I had all the answers.

          • Nature communication 2018:
            “GWAS of epigenetic aging rates in blood reveals a critical role for TERT”

            This is the published version of the preprint posted by Josh above.

  22. Agreed that intermittent rapamycin treatment should have anti-aging effects, but I see no evidence of rejuvenation. However, until something better comes along, I think rapamycin should be a major weapon in our war chest.

  23. Dobrota, I’m beginning to wonder about Russian trolling. Are you saying we should just give up our quest and submit to aging and death for the good of the economy?

    • In America, you’re doing it anyway. In America, the demagogy, which promotes aging foods, has taken hold. These are unsaturated non-hydrogenated Omega 3 and Omega 6 oils, as well as cadmium rich foods: chocolate, olives, sunflowers, fish and various superfoods. Only powerful personalities resist such demagogy. Would you eat anything harmful to make people tap their forehead?

      • You are right that cadmium probably is a significant contributor to aging. However, I disagree that we have no options to prevent bio accumulation. It is well know that lactobacillus plantarum probiotics can reduce the absorption of cadmium from the intestines. We also have chelators like NAC, garlic and a host of other sulfur containing chelators.

        • Hi Ole, are you by any chance skandinavian or are having skandinivian roots?

          I like NAC too, I really like sulfur compounds! They are very bioavailable, because they tend to be water soluble! I know that NAC is a potent mTOR inhibitor too, it can cause mouth ulcers in excessive amounts, i would guess like we see with rapamycin overdoses.

          Garlic is wonderfull too, and it also contains alot of sulfur-based molecules. I prefer (kyolic; aged garlic extract) beacuse it has underwent atleast 20 weeks in either alcohol or some glocose containing liquid and heat. This process turns all the; polysacchartides, proteins, enxymes, amono acids, gamma-glutamylcysyrines, s.allycysteine into water-soluble sulfur compounds; s-allycysteine s-allymercaprocysteine, and other sulfur base amno acids. All good stuff, all very bio-available, and bio-active. (pretty good research in vivo, and in vitro) i encourage every one to, research it.

          lactobacillus plantarum probiotics, any brand you prefer?

          I like the idea of probiotics… but are we there yet, can we get a strain to really grow in your stomach?

          just looking it up quickly;

          “The entire genome has recently been sequenced, and promoter libraries have been developed for both conditional and constitutive gene expression, adding to the utility of L. plantarum. It is also commonly employed as the indicative organism in niacin bioassay experiments, in particular, AOAC International Official Method 944.13, as it is a niacin auxotroph.[citation needed]”

          “It is also commonly employed as the indicative organism in niacin bioassay experiments, in particular, AOAC International Official Method 944.13, as it is a niacin auxotroph”

          Interesting i think?! thought expriment, what would happen if i was to supplement L. plantarum and our N-R coumpund? or just plain niacin? “Auxotrophy (Ancient Greek: αὐξάνω “to increase”; τροφή “nourishment”) is the inability of an organism to synthesize a particular organic compound required for its growth” really this could be interesing, it is late, and i need to get some sleep, Mark, Josh, Ashkay, Ole, Alan, Harold, Paul, Gabor, Cassia please Weigh in with your thoughts.

          ps. Mark i was reading in an older chemistry textbook about the Crebs cycle (2014) and Nad+/nadh synthesis and so on. There was a case study on a japanese woman who would deep dive after fish in cold water 7-8 hours aday, they did the only realiable test on brown and beige adipose tissue and she had an staggering amount! really Interesting! I am stil waiting on my inj. resv. and i will combine it with cold exposure + sirt1 activation. Will be interesting to see what happens, i will get MRI, and weekly blood test to asses it, and my generel supplement routine, every weel for 8 weeks. We most defently need to keep our BAT UP. Bat is good, and BAT is good for your stem cell population, no doubt in my mind.

          Brand

          • and in regards to your experiment with liposomal-resveratrol, did you use the actinovo one? i am not sure i am totally happy with their method of making the liposomes.

          • You are right dr brand, I am Danish.

            and thanks for the tip regarding aged garlic extract.

            As for NAC, I asked Dr. Alex Vasquez, as there has been some concern (in an earlier thread) regarding the use of NAC. The problem being that NAC is metabolized to methionine. Methionine is known as a potent mTOR activator. However, neither me or Dr. Vasquez have been able to find any studies pointing towards NAC being an mTOR activator. My own take on NAC is to use it intermittently at reduced dosages.

            I’ve used the Elixa brand of probiotics I’m not affiliated in any ways). It delivers 50x more beneficial bacterias than ordinary pro-biotics. It is meant to be taken as a 6 days course. The only thing, I’ve noticed while taking Elixa is that my skin looks much better, so something is definitely going on.

          • Hi Dr Brand

            Yes I have tried various Actinovo formulations including the Reseveratrol one.

            What concerns do you have regarding the way they make liposomes?

          • Oh and I am currently trying to encourage adipocyte differentiation to increase subcutaneous fat. I’ll know if it’s worked in a couple of month.

          • Dr. Alex Vasquez like NAC for TOR inhibition, but the recommended dose is 1600 mg x TID = 4800 mg daily. (21 doses a week) It’s cheap, can buy 1 kg powder for $38. Or take 1 dose of rapa per week.

        • That’s great news!. Why did not you still recommend it to someone who has symptoms of cadmium poisoning?. The most significant symptoms of Cd poisoning are impotence, back pain and high blood pressure, tiredness and sleep disturbances.

  24. I’d hardly base my opinions about aging on two people. Heredity has a lot to do with that – you’d have to put several identical twins on one sort of diet or another to determine how much diet influenced lifespan, but even that wouldn’t work as identical twins don’t die at the same time. If I’m not mistake, their ages at death are no different from non-twin siblings. Funny thing is, that contradicts my initial assertion that heredity has a lot to do with aging. No explanation(other than maybe I’m wrong about the ages of death of identical twin or about heredity influencing lifespan)

  25. Hi Josh, I’m interested in your interpretation of long telomeres and Horvath methylation being inversely proportional. You mention this could be like an evolved see-saw, with each capable of killing you, to make it more difficult to evolve away the consequences. You mention the report we all discussed about the SNPs associated with longer telomeres in humans also being associated with some rare cancers. Do you think that these SNPs somehow have a dual function with cancer being downstream of accelerated methylation?

  26. So the article you quote says that Harman (and this is hardly new stuff – its from 60 years ago and much of what Harman said has been disproved – radiation does not produce aging) does indicate that polyunsaturated fatty acids were linked to cancer, not aging. In the meanwhile there is no further confirmation of this. Harman had a ‘theory’ of aging – really a hypothesis, that aging was caused by free radicals, but this has never been proven, and in fact those agents which are protective of free radicals, so-called antioxidants, when used therapeutically to prevent cancer and aging, actually increased cancer rates and decreased lifespan. While they are essential fatty acids, which means the body requires but cannot make them, there is not really much evidence about how useful they are in preventing cancer or heart disease, so perhaps you are right to bring this common assumption up to be criticized — however they do reduce the occurrence of markers of inflammation, like TNF, IL-6 and C-reactive protein, in the blood. There is considerable proof that inflammation (‘inflammaging’) does result in several of the diseases of aging including cancer and heart disease. Your are correct however that there is nothing wrong with saturated fats, and reducing their consumption does not have any effect on the rate of heart attack deaths, though adding unsaturated fats may be protective. Of course everything is a matter of how much you intake. Again, population studies indicate diets high in fish and vegetables (omega-3s are producing in chloroplasts) results in longer lives (Scandinavians and Japanese being the longest-lived groups) – and the longest lived group in the US are the Seventh Day Adventists who are vegetarians (so they don’t intake cholesterol (found only in meat) or much in the way of saturated fats. So to say that saturated fats reduce aging is uncalled for – there is no evidence of that – but I guess I have to take back that ‘troll’ stuff, you do make a contribution by calling into doubt this entire saturated vs unsaturated fat mythos.

  27. Seems like there is a lot of evidence going against general hypomethylation with aging. Seems like it is really tissue dependent. One study claims there are large hypomethylated regions in PBCs.
    An Integrative Multi-scale Analysis of the Dynamic DNA Methylation Landscape in Aging

    However I found at least two very thorough investigations which claim a mild generic hypermethylation with age
    In the murine liver:
    Dietary restriction protects from age-associated DNA methylation and induces epigenetic reprogramming of lipid metabolism

    I the human thyroid, kidney, breast, glia and skin
    Distinct chromatin signatures of DNA hypomethylation in aging and cancer

    Common in all of the studies is that age related hypermethylation usually does not affect transcription, but reduced cell plasticity, whereas hypomethylation happens in those genes which are otherwise transcribed – thus tissue specific.

    Now I find most of the studies about transposable elements and aging are in Drosophila. Maybe they are not that important in humans?

    • This is no longer necessary. Can you name the most effective anti-methylation agents? I would try them and in 2 days I would tell you if it works.

  28. About the possibility that the Horvath clock is measuring just stem cell related expression. Perhaps its most important feature is that it can work across all tissues. Including post mitotic ones like the brain. Assuming that his methods were accurate and it did not include any glial cells then you have to ask why do these cell exhibit this methylation profile change over time if they have not proliferated from ‘older’ stem cells. A bit of neurogenesis has been proven to take place in the adult brain, but again, I’m going to assume here that this is only marginal to the total tissue and the accuracy of the clock.

    This still doesn’t reveal the mechanism which is driving the changes, especially in non proliferating tissue (one can say cell passage does in the rest). I was re-reading Horvath’s paper from 2013 (“DNA methylation age of human tissues and cell types”) and he puts forward the hypothesis the clock measures that the amount of work performed by a kind of epigenetic maintenance system. In the line of Harold’s idea above about energy used in metabolism.

    Another that idea has been proposed by Vince and Jim over at their blog is that cortisol, and upstream of it circadian rhythms, can drive hypomethylation across the whole genome. I guess one can imagine that the activation of certain sites, even if random at first, can trigger the activation or silencing of others. I think Josh has talked about this idea a few times and stressed the potential importance of the hypothalamus as a master regulator of circadian rhythms and by means of it of ageing.

    Personally, I find it a bit difficult to see how such a short time ‘unit’ can drive a probably much smaller set of changes over so many years. Also, one would still have to account for the ‘ticking’ of the clock in embryos or in vitro cell cultures. But I may be underestimating the capacity of the CNS to drive accurate changes over the whole body and over many years.

    I would say that there is no question that this is an ‘age clock’. Probably also the most accurate to date we have. The million dollar question is, is this is the proximal cause of ageing? Nobody knows, but I do think it is part of the picture, that is, it’s not just a byproduct.

    • First of all, you should answer my question whether the mitochondria remain old when dividing the cells or creating new ones. If it is scientifically proven that mitochondria grow older by ROS, and you do not know if young mitochondria develop, then it does not matter if DNA ages.

    • As we know methylation changes do not cause immortalized cells to age, so the only thing left in my mind is that it can impair their plasticity. Therefore its a stem cell issue.

    • It isn’t only cell division that involves the need to remethylate the genome, DNA repair probably is quite similar, so as long as cells are metabolically active errors with methylation will occur. So it is a ‘work’ thing, as Horvath suggests.

    • And Dobrota, ‘work’ involves the activity of mitochondria so yes, their efficiency or lack of it through ROS production will have a rate setting effect on the pace of aging. Therefore keeping healthy mitochondria will slow, but not prevent aging.

      • You have an amazing gift, avoiding answers and reacting illogically. You’re just playing vocabulary.
        Healthy mitochondria age as well as unhealthy and with them a whole cell whose genes adapt to the depleted energy supply.
        Even if you slow down the aging of the mitochondria, it still accumulates cadmium in the cell in the form of lipofuscin deposits and this is reflected in the aging of the organism.
        Resign to methylation. Give it up. This is not a cause but a consequence.

          • Excess cadmium has reversible and irreversible effects.
            Irreversible effects: Inhibition of ZnSOD, Fenton reaction products, Atrophy of gonads and pancreas.
            Reversible effects (if a cadmium removal method is developed in the future): improved proteosynthesis, an increase in the activity of 300 zinc and magnesium-containing enzymes.

          • Regarding organic cadmium:

            I could not post with the link, but if you google the title you can find the entire NIH abstract.

            Title: Organic cadmium complexes as proteasome inhibitors and apoptosis inducers in human breast cancer cells.

            Abstract
            Although cadmium (Cd) is a widespread environmental contaminant and human carcinogen, our studies indicate an organic Cd complex to be a potent inhibitor of proteasomal chymotrypsin-like (CT-like) activity, further capable of inducing apoptosis in a cancer cell-specific manner.

            It has been reported that the ligands indole-3-butyric acid (L1) and indole-3-propionic acid (L2) have cancer-fighting effects when tested in a rat carcinoma model.

        • You didn’t ask me anything, so how can I avoid the question?

          If you want me to answer the question you posed Adrian T, I will: mitochondria do not age intrinsically in my opinion, but mitophagy fails with age because of altered signaling from elsewhere in the cell. Mitochondria normally fuse before cell mitosis, so errors or damage to their DNA can be passed on to daughter cells.

        • Dobrata:

          Thank you for the interesting link.

          You mentioned Lipofuscin deposits.

          Some lifestyle changes or drugs or nutrients that can possibly clear lipofuscin deposits are:

          Calorie restriction vitamin E complex and increased glutathione appear to reduce or halt the production of lipofuscin.

          Piracetam appears to significantly reduce accumulation of lipofuscin in the brain tissue of rats.

          Other possible treatments:

          Centrophenoxine
          Acetyl-L-carnitine
          Ginkgo biloba
          DMAE ( this works similar to Centrophenoxine but not as well as Centropheonxine is thought to work)

          • These are all things that could have been taken 35 years ago when socialism was in us. Since I have drug containers containing only 50 pills, I have had to exclude those less effective pills. So I retain in the set only pyritinol and липоевая кислота (lipoic) from the older ones.

            Since 1985, when I have measured elevated levels of cadmium in urine, to date, the cadmium content in food has doubled. In short, the planet is infested with cadmium. Children who arrive in the world in 2050 will already grow up in 4 times higher concentrations of cadmium and will only live for 65 years.

          • I wonder, how Jeanne Calment managed to get to 122 years, despite all the cadmium contaminated chocolate she was eating?? (not mention all the lead it must have contained as well) Occasionally she was eating kilos of chocolate per week.

          • I read that she downed 2.2 kilos of chocolate /week. Since Cd is particularly high in the soil of Ecuador, maybe she avoided it. Recent testing by the same consumer group that found Cd in Ghirardelli chocolate now reports that their very intense dark chocolate has none at all. So did they remove it or is the testing method flawed, or maybe it’s much ado over nothing and you’d have to eat a ton of it to really have a detrimental effect. For now I’ll probably stick to the Baker’s unsweetened bars since it is high in flavonoids with no trace of Cd. Taste is awful.

    • my idea is that the Horvath clock mostly affetc genes that are not much expressed in normal resting state but when the cell receives growth, injury differentiation signals. there is evidence accumulating that norma cells can dedifferentiate when needed,so being a stem cell may not be a permanent state afterall.

      • Yes that is interesting and we are ‘on the same page’ GaborB. It could be the reason ‘inflammaging’ kills us – it is attempting to stimulate dedifferentiation.

        I expect that the root of aging in the human body is its limited regenerative capacity and everything else is a result either of that, or the body’s attempts to compensate.

        • There are many animals, like the axolotl that have amazing regenerative powers and still age and die. Frogs have significant regenerative powers but they too age and die.

          • That is why I said ‘in the human body.’

            Mice and rats have far more regenerative ability than us, but die much earlier from different causes.

          • The regenerative capabilities of the liver decline with aging considerably.
            For example this study claims that reduced sensitivity to growth factors might be one reason for which aged rat liver recovers slower and to a smaller volume than young rat’s liver.

            Liver Regeneration and Aging: A Current Perspective

    • We are at the best, blind men touching the elephant when it comes to understanding aging. But the study Vince and Jim quoted regarding stress drives epigenetic change makes sense to me. 85 of the 353 Horvath methylation sites were found within glucocorticoid receptor (GR) response elements (GREs), which are the sites where cortisol/GR binds to the promoter or enhancer of a gene. Stress is one big pro-aging driver that moves the clock forward I would think.

  29. ….and what makes the elephant even more “sophisticated” is that we need both physical and mental stress in the appropriate doses through hormesis.

    • Physical stresses such as exercise, UV and cold exposure , etc, are clearly helpful in controlled short bursts. Likewise perhaps for mental and emotional stresses, but these are particularly harmful when they are both extreme and chronic, and this seems to be particularly true for caregivers. This group of people have weakened immune responses to both the flu and pneumococcal vaccines as well as delayed wound healing.
      They also exhibit significant inflammation as evidenced by IL-6 levels that are 4 times higher than average ( a significant metastatic cancer risk according to a Hopkins study on IL 6 and 8). This was associated with increased mortality rates( shortened their lives by 4-8 years) and even more concerning, the effects lasted for years
      after the caregiving ceased.
      They also had shortened telomere lengths and double the incidence of severe depression.
      I bring this up as a special example of the severe damage wrought by a certain form of chronic stress. It is not clear to me however, that the usual day to day stresses are harmful , and may even represent a form of hormesis as you state.

      • Ole and Paul,
        I am in agreement. It is the extreme and/or chronic stress experience that causes damage. A lot of anecdotal reports of patients experienced stressful events just months before being diagnosed of cancer, for example.

        • And yet we get people surviving concentration camps and going on to live for 100+ years. I wonder if such people looked old before their time, but could continue in that state for many years, or whether they looked young for their chronological age at all ages.

          • It could be that the mTOR inhibition factor from the severe caloric restriction of a concentration camp actually negated the stress effect. This benefit , of course, would not be seen in care givers.
            Ole had an interesting comment about how animals quickly recover from stress events. I would assume that this is a function of the unique aspect of human consciousness wherein we are constantly in the process of utilizing our past memories while also anticipating the future, and this can make for a very stressful present moment. It is a difficult thing for us to escape.

          • I think we can consider ‘damage’ to actually be excessive anabolism, and ‘repair’ to be catabolism. So your comment about TOR inhibition protecting from stress (i.e. damage) is spot on. It would be interesting to see if lasting psychological damage or a quick recovery from mental trauma could be linked to the function of TOR and AMPK in the brain and nervous system.

      • Just to add to the confusion regarding stress:

        According to some studies, when children grow up in the same extremely abusive home environment, some will thrive, despite the abuse, and others will wither.

        One child will go on to be independent and successful in life and the other will be a basket case that is incapable of even navigating a system that can provide some assistance for them.

        Something else must be going on that causes some people to die of stress and others to fight and even thrive.

        Regarding animals, some animals do die of stress and rather quickly, while others do not.

        For example some animals such as, Deer and rabbit can die quickly from relatively minor emotional stress or injuries.

        They can even die from capture myopathy when being rehabbed by kind rescuers,

        Capture myopathy not only caused by capture but also other stressors is also known as “white muscle disease”
        ———————————–
        “White muscle disease: When the muscle is used its metabolism changes from using oxygen to using stored energy in the muscle. This leads to a build up of lactic acid which goes into the bloodstream where it changes the pH of the body and affects the heart output. If the heart doesn’t pump correct oxygen to the muscle, the muscle starts to die. Over the next week or so, the product produced by the muscle’s death damages the kidney and affects other organs.”
        ——————————————————
        Other animals like Zebra, as Ole pointed out, as well as wild boar and the water buffalo, however, can take a lot of stress and injury and still survive to fight another day.

          • I think that may be true, Paul.

            That is likely placebo drugs and sham operations are often successful.

            The people think they have been cured, so they act healthier and just live.

          • Heather you have a point, but I think the bigger point behind placebo-based cures is the connection between ‘mind’, brain and body. It is clear that the superchiasmatic nucleus in the hypothalamus of the brain, the ‘master clock’ of the body controls, for example, at least some aspects of aging.

  30. This was also an interesting study along these lines:

    The epigenetic clock and telomere length are independently associated with chronological age and mortality. R. Marioni. Int. J Epidem 2016 April 45(2). 424-32

    They looked at combined cohorts analysis and a one standard devition increase in baseline epigenetic age was linked to a 22% increased mortality risk, whereas in the same model, a one standard deviation increase in the baseline telomere length was linked to an 11% decreased mortality risk.

    • This is such a fascinating study area, but also very frustrating. We have to still be missing a vital piece of the puzzle. I just want to see a nice 3D model showing telomere loss and the associated chromatin rearrangements caused by that, and also I want to see the effects of chromatin and methylation changes far away from the telomere. I think we are being held back by what we can actually ‘see’ with our current instruments.

      • You’ll find this study on the effect of exercise on telomere length to be very comprehensive and fascinating.

        ” Telomeres, Aging, and Exercise : Guilty by Association”
        Warrick Chilton. Int J Mol Sci 2017, Dec 18(12) 2573

        • Its on my reading list, thanks Paul. Having skimmed it very briefly I have trouble accepting that exercise can increase telomere length however. I think it’s more likely exercise is similar to fasting, in that it adjusts the anabolic to catabolic balance within cells and this has beneficial effects on cell turnover and maintenance. This would have the illusion of increasing telomere length at a given age, but would not really have lengthened any telomeres.

          • The ‘illusion’ of increasing telomere length? Telomere length is measured, not inferred. It’s already been established that stress decreases telomere length and exercise is one way of dissipating stress, so it’s no that surprising.

          • They are measured but not very reliably. Leucocytes TL is very dynamic and depends on the stress on the immune system vs replacement from progenitor cells. So even before an intervention such as exercise you’d need numerous measurements to get a baseline. I think of it like measuring the depth of a bath with a big wave running up and down it. Then you have to understand the effect of the intervention, which might increase cell replacement rates, or decrease their attrition, or do lots of other things that aren’t actually increasing chromosome telomere lengths in a cell but appear to do so when looking at a population. So I think ‘illusion’ is quite an apt metaphor, albeit requiring some explanation!

          • Referring to my prior post on telomeres, aging and exercise the authors state:

            “The likelihood of measurement error in telomere research is high. Such variation is likely to render many associations non-significant or questionable at the very least”

            Something quite interesting in this paper is the finding that despite abundant associations of telomere lengths and decreased mortality rates, this association diminishes with age and telomere lengths fail to predict mortality in the very old.
            ” Paradoxically , telomere length in early childhood most accurately predicts life expectancy”
            So we should test our kids?

          • I thought that was interesting too. I suspect that children have all their telomeres around the same length, but that the old have large variations between their longest and shortest. This would mean the chance of a misleading measurement would go up markedly with age.

            So yes, telomere length at birth may well me a good predictor of max lifespan.

            Totally agree with the authors’ recommendations that telomere measurement needs standardizing.

  31. Also I find it interesting, how different species handle stress. Take e.g. a zebra, which have barely survived being hunted by a lion.In fact everyday life for a zebra can be a struggle for survival. Many humans would suffer lifelong trauma and PTSD, if we somehow had managed to escape the lions attack. However, the moment the battle for survival is over, the zebra goes on with its life. Usually when we are kids, we never worry bout anything, but as we grow older, things like war, illness, divorce etc start to trigger stress. If we could bring some of the same childish carefree attitude throughout adult life, I think a lot of unhealthy stress could be reduced.

    • This is where religious practices come into play. Buddhism uses meditation, chanting and physical repetitive engagement to train your mind not to wander. I personally find Christians’ singing and praying to be very therapeutical in reducing stress. I just have a hard time to listen to the preaching. In Ecuador, I often go to church, so I get to sing with a group of people and close my eyes to be quiet when they pray. My Spanish comprehension is not good enough to pick out all the contradictions from the pastor. 😀

    • Ole:

      Maybe the Zebra goes on with its life “as soon as the battle is over” because there is no one around to enable or coddle the Zebra.

      The Zebra does not have the luxury of making itself vulnerable to another attack from a predator by becoming lost in stress and anxiety.

      The Zebra needs to get its edge back as quickly as possible in order to survive.

  32. To me, the key question right now is how the different epigenetic clocks relate to cell senescence and the SASP, as this seems to be the closest we have to an overall proximal cause of ageing.

    In this regard, I wonder how many here are familiar with the work of Andrei Gudkov. His team published a study last year about the role of aged macrophages and the ‘SASP’. He has done a nice presentation available here: https://youtu.be/uvD5QOcf6TE.

    In it, they dubbed ‘aged’ macrophages as ‘SAMs’, for senescence associated macrophage, and showed how they appear to be the key to the appearance of the SASP, rather than senescence cells accumulation itself. In his talk he says that ‘something’ changes in macrophages over time that limits their ability to clear SC’s, so they just linger around these cells secreting cytokines that increase inflammation but never managing to fully clear them.

    He doesn’t mention the connection, but I wonder if that something that changes over time is the epigenetic profile of these macrophages. I hope his team is aware of the different epigenetic clocks and are actively looking into it.

    He’s got another talk at TED over on YouTube, where he discusses how it is ‘ancient’ viruses that ultimately cause ageing. I didn’t understand very well at the time but perhaps he was referring to several HERV and similar elements that get activated over time.

    • Horvath wrote an interesting paper on his clock and it’s relationship (or lack of one) with cellular senescence. He made some arrogant sounding statements about senescence being a bit of a side show in aging and epigenetic aging being the real deal. I expect in fact that telomere length is a good measure of the current health of the somatic tissues and epigenetic age relates to the availability of stem cells to provide replacements when needed. Hence I expect epigenetic aging of macrophage progenitors cuts off their resupply and the remainder gradually become senescent.

      • I didn’t perceive Horvath’s judgment as arrogant. It comes from the mathematics, not from the biology. Telomere length correlates with age, but the methylation clock correlates much more tightly.

        • There is obviously some overlap here but it strikes me that telomere length may be better correlated with health, since those measures which are known to improve health like smoking cessation, exercise, diet, stress control, etc., also extend telomeres. But telomeres seem to be a rather poor predictor of age since the range is so great at any given point. For instance, my father at 90 was very healthy and had a telomere length of a 20 year old, but he certainly didn’t look 20 and I suspect his epigenetic age would have been closer to his chronological one.
          On the other hand, the Horvath method appears quite accurate with regard to age, but it seems to disregard things like radiation, smoking, and diet. So at 60 you may be able to take something that reverses your epigenetic clock back to say 45, but you die at that age with lung cancer from your smoking.
          I’m suggesting here that it may be ideal, and perhaps even necessary, to achieve the best of both worlds.

        • This is the paper I was referring to Josh. ‘Epigenetic clock analyses of cellular senescence and ageing’. Horvath immortalised some human epithelium cells with telomerase and then shows how they continue to age, even though they never senesce. By age he means of course that their epigenetic methylation pattern continues to advance to what would be expected in an older person, but of course being immortalised these cells proliferate forever with nothing wrong with them. He then makes some erroneous statements about how this tallies well with the reason mice age and die, somehow forgetting they accumulate massive cellular senescence (much more than humans, though through non telomeric means).

          A better conclusions to his paper would be that his clock mirrors chronological age very well, but appears to cause no known harm to cells.

          Therefore I stand by my conclusion that he makes some arrogant statements.

          • To say that Horvath was arrogant is wrong. When you write a paper you can’t just present ‘facts’ without interpretation, and it wouldn’t be the first time that the facts were correct but the interpretation wrong.

          • Hi Mark, thanks for the paper reference. I would have said CS was central some time ago, but Gudkov’s findings highlight the fact that it is a change in the immune system rather than SC itself which is key.

            Besides, we have long hypothesized that ageing is driven by changes in gene expression. So I would say that epigenetic clocks are too precise and right where we would expect them, not to be, at least partially, drivers of the actual ageing process. Just more upstream than SC accumulation, ECM degradation, etc, that we see, possibly as a consequence of these changes.

          • Hi Adrian T,

            I was just speculating on what influence the methylation changes Horvath measures might have on aging. As I’ve stated time and again, they don’t seem to impair somatic cell proliferation, and we know it doesn’t influence cellular senescence based on the above paper, so cell plasticity (i.e. ability of stem cells to differentiate) seemed a fair bet. Maybe it does nothing and is a red herring. It might just be a byproduct of the cellular metabolism that does lead to senescence.

          • Besides you don’t need senescence to stop cells working properly. Long before they stop proliferating they start slowing down their division as well as all of their internal functions. So it could be these macrophages are just getting old because of short telomeres affecting their gene expression. I don’t think we need to hunt for a deeper explanation.

  33. Really brilliant insight in this is that if we have a more precise test for aging it would dramatically advance aging research because then people could try things and see if markers improved rather than having to wait to see if life was prolonged. This would really advance the field a lot.

    • Absolutely. I find strange though that between so many smart comments none added something on the Insilico’s AI interesting results (IMHO) and also the comparison with Hovarth’s clock as in my previous post.
      Maybe I should search if Hovarth himself mentioned something about?

      • Thank you, albedo, for the IMHO link. They listed 3 sets of blood test parameters you can use to track predicted age on their website.
        Next time I have a blood test, I will consult their parameters.

        • I agree with you Cassia. A study I would like to do (but lack time) as n=1 “data experiment” since I have data logged over many years is (i) comparing the prediction of the 3 versions at a given chronological age and (ii) see for each version the evolution in time of the delta between the chronological age and the prediction of the algorithm. Possibly (i) would give me a hint how the algorithms work in my case, meaning how the non-linearities between the various biomarkers in my own body are captured and (ii) would possibly suggest whether or not I am doing something impacting my biological aging rate (see also my reply to Mark)

          • Of course, the more people use enter data into their website, the more they can improve the parameters in the future. That’s why, I think, they put out three. However, the data are drawn from a general population and general population only lives to late 70’s, so I would think all the clocks including Horvath’s would only be good as a guide against early death. Once you are over 80, a longevity clock? I don’t know.

          • Hi Cassia. Sorry replying to me!

            I think a good clock should predict mortality risk (maybe also ideally morbidity) no matter at which chronological age. If you consider the overall increase of life span of populations and interventions which start to appear to slow and possibly reverse the aging process, having a precise clock is critical.

            Moreover, the epigenetics clock of supercentenarians is an active area of research. Horvath and Franceschi have contributed to this you might google on both.

            There are also things not really understood happening at very old age after about 115 yo (!) where the exponential increase of rate of death actually seems flattening out. Both quantity and quality of data from this extremely special population are confounding but the effect seems to be real.

          • Hi albedo,
            I agree that we need a good clock to help us measure if our regimen is working.
            Re: Mortality-rate plateau, I remember listening to Michael Rose at UC Irvine talking about aging plateaus around 95. But we are so weak by that age, our chance of dying is at 50%. Did they push it up to 115?

          • Hi Cassia. Aubrey de Grey gave a presentation last May which is on YouTube. You might be interested to check it out: “Aubrey de Grey – Limits to Human Longevity”.

      • Thanks albedo, going to read that paper ASAP. Agree with Cassia that using those blood tests as a measure of aging could be very useful in trialing interventions.

        • Thank you Mark. Please let us know your findings.

          I fully agree with you and Cassia.

          With that paper (1) I mentioned in my previous post here I also would recommend a second coupled one (2) which relates the prediction of the algorithms to the mortality risk which to me means biological age. In the second paper they say “…A Cox proportional hazards regression model was used to relate survival time to the accelerated aging group (delta >5) and slowed aging group (delta <5). Patients predicted younger their chronological age has a lower mortality risk, while patients predicted older has a higher risk. Each row represents a hazard ratio and 95% confidence interval …" In confirmation of this Insilico is also setting up a mortality predictor on their web site.

          What I also find intriguing is some of the biomarkers given as most important in chronological age prediction (e.g. higher albumin) are tantalizing related to lower cancer risks. Moreover, there is a certain degree of overlap with other AI predictors as mortalityorg and there is a clear direction in using composite biomarkers set.

          If I consider all the above, the relative stability of epigenetic markers to metabolic interventions for monitoring purposes, the easiness and zero cost (OK you need the data!), the correlation to Hovarth’s and Hannum’s epigenetic clock (as well as to other clocks) pushes me to investigate a bit more.

          (1) “Deep biomarkers of human aging: Application of deep neural networks to biomarker development”

          (2) “Population Specific Biomarkers of Human Aging: A Big Data Study Using South Korean, Canadian, and Eastern European Patient Populations”

          Sorry I am not posting URL as I am concerned the post will be rejected.

          • Wouldn’t it be great if they got enough data from simple blood tests to uncover some hereto unguessed at but critical mechanisms to aging ?

            I’ve posted this several times in the past, but you might not have read it. Protein profiling reveals consequences of lifestyle choices on predicted biological aging. Here they showed that coffee and oily fish consumption gave the greatest benefits, and smoking and soda produced the worse harm (to a measure of biological age based on protein markers in the blood).

          • Just read it. Basically eat oily fish 3 times a week, drink coffee, and exercise. Most of us probably do those things I would think. Would be interesting to evaluate fasting.
            I’m interested on your thoughts regarding sub-q fat.

          • Thank you Mark for the pointer. No I did not read it and will do. You obviously are on top of these topics since much longer than me.

          • I think the key thing Paul, is that fat is not necessarily bad, it can be a useful sink for glucose and can regulate insulin. It is only visceral fat that is inflammatory and dangerous to health. Hence why men are much more prone to cardiovascular complications than women (and indeed even cancer), because they tend to store fat in more dangerous places as they age. But if we could all keep a nice store of subq fat, we would actually be healthier than without it. Why do we lose it at such a young age, that is the question?

          • That’s interesting and certainly not the conventional thinking. So do you think that the more pear shaped women are protected whereas apple shaped men with their subcutaneous abdominal fat are at risk?

          • Any kind of subq fat is good, including abdominal. But visceral fat around the organs which tends to be seen as belly fat is highly dangerous. The interesting thing about the paper ‘Pioglitazone Increases the Proportion of Small Cells in Human abdominal Subcutaneous adipose Tissue’, is that subq fat increased but visceral fat decreased. Blood glucose went down.

          • It is my understanding that visceral fat feels firm whereas sub q fat is soft and mushy around the waist. So for the bodybuilders who reduce their body fat to 5%, Maybe not so good?

          • Progess is being made in this area Paul, just look at: “Phenotypes of prediabetes and stratification of cardiometabolic risk.”

            Dr Norbert Stefan have wrote quite a bit about this.

  34. Hi Paul,

    Regarding my previous research on subq fat, here are the choicest papers.

    ‘Adiponectin Receptor as a Key Player in Healthy Longevity and Obesity-Related Diseases’.

    And here is where they tried it out.

    ”Pioglitazone Increases the Proportion of Small Cells in Human abdominal Subcutaneous adipose Tissue’.

    But there is a small risk of bladder cancer, as per the below.

    ‘Pioglitazone use and risk of bladder cancer: population based cohort study’.
    ‘Deciphering the Roles of Thiazolidinediones and PPAR𝛾 in Bladder Cancer’.

    For a while I thought I could use magnolol instead of pioglitazone, but found it ineffective. So I am now carrying out 3 month trial as per the 2nd paper I referenced above.

    • Pioglitazone have had me interest for long, i am looking forward to hear about your results or lack thereof. I would be interested in seeing if it can be used to fasten the creation of Brown/beige fat deposits. (high fat meals, mct + cold exposure + sirtuin-1 activation = brown lots of brown fat?)

      You asked for a study that showed sirtuin activation = brown fat, i found a very beautiful one, i will post title, i just cant find it ATM. but i got it saved somewhere. 🙂

      I dont know why it effects the liver, i would expect the Pioglitazone to activate PARP-y and thereby reducing one’s NAD+ pool. wich might lead to increased cancer in the bladder? (i really dont know, i am just theorizing at this point.)

      • If you have any ideas about how I can test for Brown fat Dr Brand, please let me know. I know new fat cells are being generated as the health check machine i have nearby has increased its estimatation of my fat % from 14% to 37%! I’ve used it lots of times over the last 8 years as it’s free, and never been out of the 12-17% range. I’ve not put on much weight though (within normal variation), although my emaciated cheeks are starting to look a little fuller!

        Paul, I think subq fat is lost from childhood onwards. If you look at someone in the 50 or 60s who is getting out of shape due to age, even with lots of fat on their belly and chests, their face often still looks thin. I expect bodybuilders don’t lose all their remaining subq fat, just shrink the cells right down, though I could be wrong.

        • That sounds wild, but 14% to 37% with you not putting on much weight sounds a bit… contradicting unless you changed body composition too? you know lost a lot of muscle and gained a lot of fat? do you think the Pioglitazone, might be tricking the machine somehow? mmh…

          Well in my generel, organic, and biochemistry book, there is this case study with a KOREAN ( last i mentioned it, i said she was japanese, not true) woman that lived by diving in icecold waters for 6-7 hours a day, with an amazing amount of brown fat, they clearly but it as a consequence of cold-adaption.

          But i quote “However, this thermogenesis literally burns up most of the brown fat tissue, and adults typically have so little brown fat that it can be found only by using a special technique called, Thermography, which detects temperature differences troughout the body”

          I dont know of any other ways to do it, but google thermography and search for pictures, it actually seems like a reasonable way to do it.

          Have you encountered any side effects using it? and do you continue with rapamycin when on Pioglitazone?

          • Before China became rich, government regulation only allow north of Yangtze River to heat in the winter. I used to live in Shanghai, which borders Yangtze River on the south side. Every winter I would gain 12 pounds from the summer 95 pounds. Literally my thighs expanded in the winter time. Last 30 years I have lived in Canada, so no more weight gain in the winter time.
            According to Wikipedia, female hormone is at play for women to store fat in buttocks, hip and thighs. After menopause, fat migrate to abdomen due to loss of female hormone.
            I have kept my weight at 95 pounds and have not noticed any fat gain in my abdomen. But I did notice fat loss in my thighs.

          • I’ve had no side effects but I’ve only been doing it for a month. I’m not sure why the machine is measuring such a high fat value as that would make me obese, and in fact my body composition hasn’t changed noticeably . I only mention the machine measurement as it was unexpected and suggests something is happening. And my face does look a little fuller. I’ve tried both with and without rapamycin as I thought it might stop the desired fat cell proliferation, but it made no difference and is one of the reasons I started to reassess what rapamycin is actually doing. I now think that rapamycin does not stop proliferation at all when taken intermittently, and may even increase it in Vivo, because it probably saves lots of temporarily arrested cells from permanent senescence. I’m going to continue this experiment for another 1-2 months and then do a blood test to check everything’s okay and hopefully I’ll be slightly fatter (but not in the belly) but have a healthy low glucose level.

          • So where do you think that all of that additional fat has gone and are you sure it isn’t visceral?
            That ‘s an interesting thought on rapamycin. I also notice that my hair and nails grow normally with it. Could it slow growth and development without affecting proliferation?

          • Hi Paul, well base on the study they didn’t put any weight on around the waist, so I’m guessing none of it is visceral. Expect new fat cells are very small, so might not be that noticeable.

          • I think rapamycin may even improve proliferation if taken intermittently. Perhaps knocking down TOR for a short time has none of the downsides and only upside.

  35. I’ve seen 190-191 on some of my peak efforts on a bike. Very difficult to reach unless you are chasing someone or being chased. Before dosing low 180’s would be the highest I would see. Low 180’s more common now. 57 years old so that’s not too bad. I was just wondering if anyone else is seeing something similar.

    • That’s very impressive and you might be on to something. The loss of MHR with aging is believed to be secondary to a loss of beta adrenergic response of the sinoatrial node of the right atrium of the heart. It’s universally observed. Could rapamycin actually alter it? That would be amazing.

      • I took a break from rapamycin for a month and I definitely noticed the difference in terms of reduced athletic performance and greater tiredness. I wonder what the mechanism could be? Could it be as simple as reduced stress on proteostasis in many cells across the body?

        • I really like to hear all of these nice first hand reports, i was afraid that rapamycin would only “slow down” one’s usage of stem cell’s, so one would have a bigger pool of fresh one’s for longer time, however when i hear things like, marks, pauls, alan’s, experiences i can’t help but think there is more to it.

          One mechanism i can think of mark, is “intense” mitochondrial biogenesis, we know fasting does this after 24 hours + could rapamycin “fake” a 24 fast? perhaps…

          I wanted to share this too, we discussed sulforaphane ealier, and based on this study it might needs to be cycled:

          “Sulforaphane is a Nrf2-independent inhibitor of mitochondrial fission”

          • Dr. Brand, I have to tell you, loath though I am to, that I too have been taking rapamycin and it seems to have had startling effects. Prior to taking it, I could hardly walk, walked like an old man which is apposite as I’m in my seventies. I also had knee problems and intestinal problems that seemed to prevent me from eating wheat gluten, beans, eggplant, tomatoes etc. Now I have none of those problems. My friends tell me that might be because I’m in India now, and their yogurt is particularly beneficial, I don’t know – I can’t run a controlled study of all the factors, but after a month’s absence, I’m back on rapamycin (and back in India). No, I don’t think it’s ‘the’ answer to aging – but it seems a wonderful stop-gap measure. I wish I could get my physician back home to prescribe it.

          • I imagine that rapamycin works through a variety of different mechanisms to improve one’s health. When I started it about 7 months ago , I did so with much trepidation because I felt ” fine”, but I still saw a very significant improvement after 3 months of usage. The primary effect was on stamina and endurance, areas in which I was slipping , but it was so gradual that I didn’t really notice it. I have been convinced that this was due to an increased cardiac output secondary to an improved stroke volume, but now after LO’s comment on his increased max heart rate.
            The max HR invariably drops with aging, so much so that it greatly impairs endurance and often leads to the weakened and feeble state of old age. So I now ascribe my increase in cardiac output to a probable increase in max heart rate.
            That in and of itself can make you feel years younger.

          • Hi Harold, How long have you been taking Rapamycin and at what dosage?
            You are in India, it’s a wonderful country to get any kind of generic drugs without prescription.

          • So true Cassia, especially easy when you’re working with pharmacologists! I’m taking three mg, every week (I know there’s accumulation, I’m counting on it) – I only dose for about 8 weeks at a time then take a month break, (honestly this is the second round for me ever, so it’s not like a habit). I’ve been surprised that it is so effective. Hopefully we’re doing something that will make rapamycin ‘old hat’. We want to reverse aging, not delay it. I’m surprised that rapamycin seems to have reversed some aging conditions, it’s a sort of clue I think.

          • We could measure Heart Rate Variability too, apparently that’s a great measure of mortality. If rapamycin increases that, then it will be for sure it’s affecting the cardiovascular system directly.

            I think even Alan Green is at a loss for why rapamycin doesn’t just slow aging. I am 40 and I notice it’s effects, Crazy!

          • HR variability is a good measure but nothing surpasses cardiac output. It is the best predictor of survival post-MI and bypass surgery ( measured as stroke volume on ultrasound) of any other parameter. It is also very linked to endurance and overall well being .As we age the deterioration of SA node function and maximal heart rate is invariable and in my mind devastating, especially with exercise.
            Just consider a 20 year old who can get his HR up to 200 with exercise, and couple that with the stroke volume of a 20 y/o, and you have an enormous CO. They also exhibit very short recovery times. I think that it’s likely that rapamycin increases maximal heart rate and shortens recovery times. It also seems to heal, restore, and rejuvenate. I wouldn’t now stop it for even a week ( though I’m not certain how it’s doing all of it)
            To my mind, at the present time, this is by far the best anti-aging weapon that we have and I think that our efforts should be to find things to work with it in synergy.

  36. Anyone heard of these guys?

    inflazome(dot)com

    Apparently they can block specific inflammasomes and thereby safetly reduce inflammation.

    • oh my this posting system is horrible, i replied to severel people, but it just wont let me post. The above look very interesting, it would be amazing to block excess inflammation, i will read the studies, i am curious about their approach. Blocking it at the inflammasome level would be amazing.

      Mark, how do you feel Methylene blue? It has some really interesting connection to our mito’s. (Cytochrome c oxidase)

      and see the study: “Methylene blue inhibits NLRP3, NLRC4, AIM2, and non-canonical inflammasome activation”

      Ole, i read alot about elixa about a half year ago, i found them to be the best product on the market then. (i have not tried it my self yet, it is quite expensive). But it is a good indicator that you see skin improvement, the skin and stomach as you surely know have a really high turnoverrate so i would expect to see the quickest results in those areas.

  37. A review paper “Longevity, Aging and Rapamycin” by Dan Ehninger details what Rapamycin does or doesn’t do. Very interesting read.
    When it comes to gene expression, there is a huge difference between female and male mice. Long term Rapamycin treatment led to 2504 gene up-regulated & 2257 gene down-regulated in female as compared to only 159 up-regulated & 129 down-regulated in male mice.
    Another difference between female and male mice is in the concentration of the Rapamycin. In mice treated with 4.7, 14 or 42 ppm, all three doses extend life span in female, but only the two higher doses extend the life span of male.

Leave a Reply

Your email address will not be published. Required fields are marked *