2020 February 18
I found this talk very interesting, informative, exciting, and inspiring. Not because Sinclair is suggesting that we may live forever (he is not), but because he tells us it is possible to age healthily and, perhaps in the near future, it may be possible to rejuvenate some parts or organs of the human body.
Recently I re-viewed his talk for the purpose of marking out where the key points are made, for future reference. The result is posted at the end of this writeup.
Although Sinclair's is a general talk, many lay persons may still find it not so easy to follow, especially those who have not previously heard of epigenome, a cellular system that is key to appreciate Sinclair's talk.
Here, strictly for the lay person, I provide some background on epigenome and related topics that are crucial to Sinclair's discovery. These include stem cell, the four Yamanaka factors, and "reprogramming".
Fortuitously, my work (far removed from Sinclair's and many rungs below his in terms of merit) in the last ten years brought me into contact with these topics.
Stem cell and cell differentiation
Cell differentiation leads to cell identity; a cell with a certain identity - for instance, skin cell, nerve cell, liver cell - performs a certain set of biological functions.
Human life starts with in single fertilized egg, which grows and develops by many stages of division and differentiation, ultimately becoming the full body. Here, we can understand differentiation as specialization. Adult cells are finished with differentiation and are completely specialized.
A stem cell is a cell that has not reach the end of specialization; it retains a degree of capability to further differentiate. There are many levels of such capabilities, hence a hierarchy of stem cell. The mother of all stem cells is the fertilized egg, then, at one level lower, the embryonic stem cell.
Almost all cells in a born baby are already adult cells: nerve cells, lung cells, liver cells, skin cells. Each type, with its unique cell identity, does one set of things and not any other. Yet they all came from the single fertilized egg and all have the same DNA as the egg. What sets cells with different identities apart?
Think of the DNA in a cell as the hard disc of a computer (11:50). Just like the DNA that carries the code of many genes which, when expressed, make proteins that are essentially nano-machines in the cell, the computer hard disc contains many subroutines (like apps on your cell phone) that allow the user to do many different things.
For a certain task the user opens one subroutine or a set of subroutines, leaving others alone. Similarly, in cells of a specific identity only genes that code proteins needed for the cells' designated task are expressed.
Yet this analogy is not complete. A user doing one task can change her mind at any point and open another subroutine to do another, possibly quite different task. A specialized cell cannot. A higher level of analogy is needed.
Consider a big corporation making and selling a number of products. All its intellectual properties are stored in a corporate hard disc. The corporate's various departments, or branches, each responsible for one product, all have access to this corporate hard disc. Not in its entirety, however. Each unit can access only that part of the hard disc relevant to the product it is responsible for, the rest of the hard disc, as far as that unit is concerned, is locked. This assures a task-specific unit in the corporation can do only what it is assigned to do, not anything else.
In an exact analogy, in a cell with a specific identity, that part of the DNA not needed for its assigned task is locked. This ensures that throughout the entire life of an adult cell, it can do one and only one job.
In the universe that is the human body (or the body of any multicell species), the political system is extreme authoritarianism, and cells are not individuals with free wills, but rather robots destined to one and only one job its entire life.
Presumably, the locking phenomenon was developed during the long course of evolution to give a living organism an extremely high degree of stability to each of its many diverse parts.
The epigenome locks up part of the DNA to give a cell its identity
Epigenome (表觀基因體) (10:10)) is the name for all things that affect the function of a cell that do not involve alterations in the DNA. The epigenome has two main functions: packaging the DNA and the locking/unlocking of certain parts of it to give a cell its identity.
When stretched to full length the human genome is about 1.8 meters long; when packaged into chromosomes each is about one to two micrometers long.
Unlike the genome information coded in the DNA, which is digital and precise, epigenome information is analog (11:09) and probabilistic. Namely, there is not a single key that locks or unlocks a specific, precise site of a DNA. Rather, series of chemical actions, however affected, would lock, with a probability, a general area of a DNA.
When the DNA in a cell is partly locked down to give the cell its identity, the DNA exhibits an approximate locking signature associated with the cell identity. The epigenomic locking signature is inheritable; identical twins share the same genotype but may have different locking signatures, hence different phenotypes.
At 10:10 of Sinclair's talk there is a cartoon showing what the epigenome looks like: the genome (or DNA) wrapped around spools. Each spool is composed of four specialized proteins which, through chemical interactions with the DNA wrapped around it, perform the locking function.
Yamanaka and the reprogram of differentiated adult cell
Differentiated adult cells can be "reprogrammed" into an earlier stem-cell state
As mentioned above, when a cell is fully differentiated - skin cell, liver cell, nerve cell, bone cell, etc. - it acquires a distinct cell identity with its DNA partly locked up to ensure that it does one and only one kind of job.
In 2006 and 2007, Yamanaka Shin'ji (山中伸彌) and his group published two papers in the journal Cell reporting that they were able to reprogram adult cells taken from mouse  and human  into induced pluripotent stem cells that have the differentiating capability one level below the omnipotent embryonic stem cell.
Yamanaka showed that artificially importing four specific genes, now known as the four Yamanaka factors (22:32), which are normally locked in adult cells, into an adult cell triggers a process in which the cells would, over many divisions (as many as 60 generations), gradually fully unlock the epigenome, and this turns the cell into a stem cell that has the capability to differentiate into a variety of adult cells distinct from what it was before.
The publication of these two bombshells led to Yamanaka, at age 50, being awarded the Nobel Prize in Physiology or Medicine in 2012 (22:04).
Sinclair's Information theory of Aging
Cell aging is caused by loss of epigenetic information (8:33)
Just like the genome is an information system whose contents (mostly genes) define species, within a species the epigenome is an information system whose contents (the gene locking signatures) define cell identity. Loss of genome information could lead to failed embryo or a deformed or diseased individual, whereas loss of epigenome information - this is how Sinclair likes to call it - would lead to loss of cell identity.
Sinclair's Information theory of Aging says loss of epigenome information in cells is a cause of aging. In his words (12:54) "What happens is that cells lose their identities ... a nerve cell in an older person is no longer fully a nerve cell,... and may be partly a skin cell."
Here, as far as cell identity is concerned, losing epigenetic information means the pattern of DNA locking is disturbed. Namely, some parts of DNA that should be locked are not, and parts that are not supposed to be locked are. The result is that a cell may no longer properly do what it is supposed to do, while at the same time do things poorly that it is not supposed to.
In experiments with mice, Sinclair's group has found a method to cause the epigenome to lose its information resulting in a mouse to biologically age at a rate much faster than it would age chronologically.
How epigenome information is lost
Repeated double-strand breakage of DNA causes loss of epigenome information
Recall that the DNA is a double-stranded helix. It was known that when a DNA suffers a double-strand break, the epigenome is triggered to move a machinery (a set of macromolecules) to the breaking site on the DNA to repair the breakage, after which the machinery returns to its normal site within the epigenome infrastructure.
However, once in a while this return action is not exactly executed and as a result the epigenome suffers a slight damage. The method used by the Sinclair group to disrupt the epigenome is to make repeated random double-strand breakages of the DNA, and the accumulated damage sustained by the epigenome ultimately causes sufficient epigenomic information to be lost for a cell to lose its identity.
In two papers, Sinclair's group showed that this epigenome disrupting procedure results changes in a mouse that match genotypically (DNA and genes)  and phenotypically (appearance, and mental and physical capabilities)  changes caused by normal aging.
Sinclair's "age reversal" experiments
The vision of mice blinded by glaucoma or old age is restored to its youthful state
If aging is caused by a cell losing its identity, or, equivalently, by the cell having a disrupted DNA locking signature, can aging be reversed by righting the locking signature? This seems a hopeless task, because there is an infinite way a DNA locking signature may be fouled up, and even more hopeless when what cell-specific correct locking signatures are not known.
By restoring vision in a mouse with artificially induced glaucoma, and in normal old mice with artificially crushed optic nerve  (26:34), Sinclair's group showed that age reversal is possible at least in some specific cases.
Sinclair's group achieved these feats by using three of the four Yamanaka factors to partially reprogram the aged/damaged cells to their youthful/healthy states (23:30). Sinclair believes this "age reversal" is achievable because "old tissues retain a faithful record of youthful epigenetic information that can be accessed for functional age reversal" (27:08).
A possible realization of this abstract notion might be as follows: the correct locking signature, after being in place for a longtime, leaves a pattern "groove" that could be more easily returned to after the pattern has been disturbed. Presumably, by performing the reprograming slowly, the disrupted epigenome was gently goaded back into its youthful state.
"The best thing to do is to eat less often and exercise"
This was Sinclair's answer when he asked himself the rhetorical question: "What if one doesn't want to see a doctor for advice or take supplements?" (30:05)
Sinclair's talk ends at 34:50. After that is Q&A period. The Q&A's are interesting but rather technical (see, however, below).
Following are some related issues.
Why did Sinclair use only three Yamanaka factors, not four, in his age reversal reprogramming?
Sinclair et al. only wanted to reprogram the cell to its youthful adult state, not to its very early stem cell state, as Yamanaka did. The fourth Yamanaka factor left out in Sinclair reprogramming procedure is oncogenic, namely, it promotes cancer growth.
It turns out induced stem cells when left undifferentiated have a significant chance of turning tumorous. One reason is that cancer cells and stem cells share a crucial property, immortality: the mechanism leading to cell death is turned off in both. (The four Yamanaka factors, actually for genes, are Oct3/4, Sox2, Klf4, c-Myc. The cancer causing one not used by Sinclair is c-Myc.)
How original is Sinclair's work reported here?
After Yamanaka showed how to reprogram a cell, many people have used the technique to do many things, including using the four Yamanaka factors to rejuvenating cells.
The important messages reported in Sinclair's talk are based on original work done by his group on mice: his Information Theory of Aging; demonstrating that double-strand breakage of DNA leads to disruption of the epigenome that can cause a cells in the mouse to lose their identities, such that a chronologically young mouse would show signs, physiologically  and molecularly , very similar to those of an old mouse; and his "age reversal" experiments showing that using three Yamanaka factors to reprogram aged/damaged cells can restore the vision of age/glaucoma affected mice .
Mice are not exactly human, but biologically the two species are awfully close and what Sinclair et al. see in mice has a good chance to be true in human, too. But that is not a given. Sinclair has had many of his papers published in first tier journals including Cell and Nature, quite likely one or more of these three preprints will be, too.
How are Sinclair's work/claim received?
Personally, I think the works reported in [3, 4, 5] are solid, in fact first class. Sinclair's many public talks, most given to promote his book (same title as the talk), seem to be raising some eyebrows [Has Harvard's David Sinclair Found the Fountain of Youth?] as well winning kudos [Can David Sinclair cure old age?].
It could be that some of Sinclair's distractors object to his hyperbole, such as naming his findings a theory, and using non-biological metaphor such as Shannon's information theory to describe how the epigenome sets cell identity, or to his too-strong advocacy that aging is a medical condition that can be treated, but not something that is natural and should be left alone.
Sinclair's Shannon Information metaphor
(This section a is a bit more technical than others) Sinclair believes that underlying the success of his "age reversal" experiments in restoring the vision of old/glaucoma affected mice is the existence of an epigenetic correction system in the cell similar to the Shannon's correction system in communication (20:44) that allows for arbitrary low transmission error, making today's IT industry possible.
My feeling is that this belief is misplaced. This notion in fact seems to contradict what Sinclair says about epigenome information: it is analogue (11:09), not digital, whereas Shannon's theory deals with digital information.
I also feel that evolution would not have produced such a correction system, not because it is elaborate, but because there was no need for it. During development a cell acquire its identity through differentiation, and functional stability required that it stays with its designate identity through its entire life.
More likely Sinclair's mouse vision experiments succeeded because of the "groove" scenario mentioned earlier.
The lack of a correction system advocated by Sinclair, if true, can be compensated, say, by taking records of the epigenome locking signatures of the various cell types of a youthful individual, to be used for Sinclair-type cell reprogramming when that individual is older.
David Sinclair supplements
You can find out more about the three "David Sinclair supplements" here. According to Sinclair, these supplements "turn on the longevity pathway", or, more precisely in Sinclair's understanding, protect the epigenome information that defines cell identity from disruption.
Many studies, Sinclair's among them, have shown that the longevity pathway is turned on when the body is under caloric stress, a state naturally attained by fasting. Taking Sinclair's supplements is a "stressless" way of turning on the pathways, but without getting the other benefits fasting, such as weight control.
Some key statements in Lifespan: Why We Age and Why We Don't Have to
2:22 - "Right now aging is not considered a medical condition ... Let's thing about it, why don't we?"
6:08 - We have a new understanding of what causes aging ... actually resetting the aging clock of the body.
8:33 - Aging is a loss of epigenetic information ... Information theory of Aging (ITA)
8:50 - Genetic and epigenetic
10:10-38 - How epigenome looks and works
10:50 - It is epigenome that give cells their identity
11:09 - Epigenome is analogue information
11:50 One way of thinking of the epigenome is that it is the software of our cell, and the genome is the computer
12:54 - What happens is that cells lose their identities ... a nerve cell in an older person is no longer fully a nerve cell, ... and may be partly a skin cell.
13:28 - Question is though, can we slow this down? Can we reset the system? Is there a reboot? Is there a backup hard drive of this early set up, that we can access, and restore the early structure? ... I believe it is possible.
16:16 - Ten years of work ... is the discovery that broken chromosomes disrupt the structure of those hose reels (of epigenome), and cells start to lose their identities and don’t function very well. And the ultimate outcome off cells losing their identities is aging. [3, 4]
17:36 - We can now measure age with great accuracy. ... This 100% quantitative. ... Measure
DNA “Methylome”. Measure which of the letter C’s, out of A, T, C, G, have a methyl group.
20:44 - Claude Shannon’s theory of information and “Schematic diagram of a correction system”.
21:38 - The transmitter in our fertilized egg, the receiver is our body in the future... We lose a lot of the information (in the course of our lives); we succumb to entropy.
22:04 - Man on the left (Yamanaka Shin'ji) won the Nobel Prize for learning how to make an adult cell into a stem cell. [1, 2]
22:32 - The four Yamanaka factors are called O, M, K, and M for short,
23:30 - "I think we finally found how to tap into the observer and reset our biological age, using the (three) Yamanaka factors (not including M)". 
23:48 - Give credit to a student in our lab, Wanqing Lu (first author of ).
24:38-57 - The "Optic Nerve Crush" experiment.
26:34 - "We can reprogram the retina of an old mouse and make it see just like a young mouse again".
27:08 - "We found the communicating device (the "backup hard drive of this early set up" mentioned at 13:28) back to the observer at least. There are a couple of enzymes, TED1 and TED2, that remove the chemical groups off the DNA group as part of that reset process".
28:40 - "NMN ... turns on the longevity pathway that we've worked on for many years".
30:05 - "What can you do if you don't want to go to the doctor and ask for metformin?" ...The best thing to do is to eat less often.... (and) Exercise ...
34:50 - Talk ends. Beginning of question period.
35:00 - Q: Why metformin, diabetes drug? A: Insulin is key.
37:26 - Q: What about the conventional hallmarks of aging. A: The proposed Information theory of aging (ITA) is upstream of the causes of those eight or nine hallmarks.
39:21 - Is oxidative damage (stress) upstream or downstream, or separate from ITA? Q: It is both. It's part of the positive feedback loop. Oxidative stress can cause DNA break. Also cosmic rays, CT scans, X-rays... A lot of vitamin C and mega doses of vitamin E can blunt the effects of a healthy diet and exercise.
41:40 - Q: What about non-nuclear oxidative damage; collagen or even extracellular things. A: Best way to tune on autophagy, chaperon mediated autophagy (to clean out damaging, unwanted proteins) is to fast for two days ...
 Takahashi, K.; Yamanaka, S. (2006). "Induction of Pluripotent Stem Cells from Mouse Embryonic and Adult Fibroblast Cultures by Defined Factors". Cell. 126 (4): 663-676. https://doi.org/10.1016/j.cell.2006.07.024. PMID 16904174.
 Takahashi, K.; Tanabe, K.; Ohnuki, M.; Narita, M.; Ichisaka, T.; Tomoda, K.; Yamanaka, S. (2007). "Induction of Pluripotent Stem Cells from Adult Human Fibroblasts by Defined Factors". Cell. 131 (5): 861-872. https://doi.org/10.1016/j.cell.2007.11.019. PMID 18035408.
 Hayano et al. DNA Break-Induced Epigenetic Drift as a Cause of Mammalian Aging. bioRxiv 2019Oct21, http://dx.doi.org/10.1101/808659
 Yang et al. Erosion of the Epigenetic Landscape and Loss of Cellular Identity as a Cause of Aging in Mammals. bioRxiv 2019Oct19, http://dx.doi.org/10.1101/808642
 Lu et al. Reversal of ageing- and injury-induced vision loss by Tet-dependent epigenetic reprogramming. bioRxiv 2019Jul31, http://dx.doi.org/10.1101/710210
HC Lee, February 2020, Taoyuan, Taiwan