There is great promise in 2020 that we might be able to make our bodies young without having to explicitly repair molecular damage, but just by changing the signaling environment.
Do we need to add signals that say “young” or remove signals that say “old”?
Does infusion of biochemical signals from young blood plasma rejuvenate tissues of an old animal? Or are there dissolved signal proteins in old animals that must be removed?
For a decade, Irena and Mike Conboy have been telling us removal of bad actors is more important. But just last month, Harold Katcher reported spectacular success by infusing a plasma fraction while taking away nothing. Then, last week, the Conboys came back with a demonstration of the rejuvenating power of simple dilution. [Link to their new paper]
They simply replaced half of the blood plasma in 2-year-old mice with a saline solution containing 5% albumin. What is albumin? Blood plasma is chock full of dissolved proteins, about 10% by weight. About half of these are termed albumin. Albumin is the generic portion. It doesn’t change through the lifetime. It doesn’t carry information by itself. But albumin transports nutrients and minerals through the body.
The Conboys took care to show that albumin has no rejuvenation power on its own, and had nothing to do with their experimental results. Rather, they had to replenish albumin in diluting blood, because the animals would be sickened if half their albumin were removed. Replacing the albumin in a transfusion is akin to replacing the volume of water or maintaining the salinity.
In preparation for this experiment, the Conboys have invested years in miniaturizing the technology for blood transfusions, so that mice can be subjected to the same procedures that are commonplace in human hospitals.
The Conboy lab replaced 50% of mouse blood plasma. They got spectacular results with a single treatment, based on a lucky guess. They have not yet experimented with 30% or 70%. They don’t know yet how long the treatment will last and how often it needs to be repeated.
Evidence of rejuvenation
As with previous papers from the Conboy lab, the group focused on repair and stem cell activity as evidence of a more youthful state. Three separate tissue samples were taken from liver, muscle, and brain.
“Muscle repair was improved, fibrosis was attenuated, and inhibition of myogenic proliferation was switched to enhancement; liver adiposity and fibrosis were reduced; and hippocampal neurogenesis was increased.”
- They measured nerve growth factors in the brain, and detected a more robust response, typical of young mice
- They lacerated muscles and showed repair rates typical of much younger animals
- They examined microscope slides of liver tissue, and showed that it is less fatty and striated than is typical of older mice
Figure 2. Rejuvenation of adult myogenesis, and albumin-independent effects of TPE. One day after the NBE, muscle was injured at two sites per TA by cardiotoxin; 5 days later muscle was isolated and cryosectioned at 10 µm. (A) Representative H&E and eMyHC IF images of the injury site. Scale bar = 50 µm. (B) Regenerative index: the number of centrally nucleated myofibers per total nuclei. OO vs.ONBE p = 0.000001, YY vs ONBE non-significant p = 0.4014; Fibrotic index: white devoid of myofibers areas. OO vs ONBE p = 0.000048, YY vs YNBE non-significant p = 0.1712. Minimal Feret diameter of eMyHC+ myofibers is normalized to the mean of YY . OO vs. ONBE p=3.04346E-05, YY vs. YNBE p=0.009. Data-points are TA injury sites of 4-5 YNBE and 5 ONBE animals. Young and Old levels (detailed in Supplementary Figure 1) are dashed lines. Representative images for YY versus YNBE cohorts are shown in Supplementary Figure 6. (C) Automated microscopy quantification of HSA dose response, as fold difference in BrdU+ cells from OPTI-MEM alone (0 HSA). There was no enhancement of myogenic proliferation at 1-16% HSA. N=6. (D) Meta-Express quantification of BrdU+ cells by automated high throughput microscopy for myoblasts cultured with 4% PreTPE versus PostTPE serum and (E) for these cells cultured with 4% of each: PreTPE serum + HSA or PostTPE serum + HSA. Significant increase in BrdU positive cells is detected in every subject 1, 2, 3, and 4 for TPE-treated serum (p=0.011, <0.0001, <0.0001, 0.0039, respectively), as well as for TPE-treated serum when 4%HSA is present (p<0.0001, <0.0001, <0.0001, =0.009 respectively). N=6. (F) Scatter plot with Means and SEM of all Pre-TPE, Post-TPE, +/- HSA cohorts shows significant improvement in proliferation in Pre TPE as compared to and Post TPE cohorts (p*=0.033), as well as Pre+HSA and Post+HSA cohorts (p*=0.0116). In contrast, no significant change was observed when comparing Pre with Pre+HSA (p=0.744) or Post with Post+HSA (p=0.9733). N=4 subjects X 6 independent assays for each, at each condition. (G) Representative BrdU IF and Hoechst staining in sub-regions of one of the 9 sites that were captured by the automated microscopy. Blood serum from old individuals diminished myogenic cell proliferation with very few BrdU+ cells being visible (illustrated by one positive cell in Pre-TPE and arrowhead pointing to the corresponding nucleus); TPE abrogated this inhibition but HSA did not have a discernable effect.
What’s missing? They did not test any measures of physical or cognitive performance at the level of the organism.
- Evidence of behavioral changes (learning and memory, endurance, strength)
- Inflammatory markers
- Blood lipids
- Methylation clock (Horvath, UCLA) or proteomic clock (Lehallier, Stanford)
Some of this is planned for future research. Mike and Irina plan to submit tissue samples for analysis by the Horvath mouse methylation clock.
I am a committed enthusiast for the methylation and proteomic clocks that are the best surrogates we have for aging. These technologies can tell us whether anti-aging interventions have been effective without having to wait for animals (or humans) to die before reporting results. But the Conboys still regard these technologies as unproven, and they bristle at the word “clock”. The closest they come is to catalog the entire proteome of treated mice, comparing it to untreated young and old mice.
Multi-dimensional t-SNE analyses and Heatmapping of these data revealed that the ONBE proteome became significantly different from OO and regained some similarities to the YY proteome. Supplementary Figure 4 confirms the statistical significance of this comparative proteomics through Power Analysis, and shows the YY vs. OO Heatmap, where the age-specific differences are less pronounced than those between OO vs. ONBE, again emphasizing the robust effect of NBE on the molecular composition of the systemic milieu.
Translation: As controls, they had mice that underwent plasma exchange with mice of similar age. YY were young, positive controls, and OO were old, negative controls. Treated mice were ONBE=”Old—Neutral Blood Exchange”. Rather than relying on “clock” algorithms that compute an age from the proteome, they compared the entire proteomes of test animals with those of old and young animals, and foud that they resembled the young animals more closely.
Aging and epigenetics
I was an early advocate of the theory that aging is driven primarily by changes in epigenetics. Other proponents include Johnson, Rando, and Horvath. This theory is now mainstream, though its acceptance is far from universal. (The main reason people have difficulty with the idea is the question, “why would the body evolve to destroy itself?” I present a comprehensive answer in my popular book and my academic book.)
On the face of it, the new Conboy result is powerful evidence for the epigenetic theory. They have shown that there are proteins in the blood that actively retard growth and healing. Remove half theses proteins and the animals are able to grow youthful tissues and to heal better. The obvious conclusion is that, with age, there are signaling changes in the blood that weaken the animal and inhibit repair.
There are, however, other ways to interpret the changes. Aubrey de Grey has said (personal communication)
“When everything in the blood except the cells and the albumin is replaced by water, the body will definitely respond by synthesising and secreting everything that it detects a shortage of, whereas the bad stuff will not be so rapidly replaced, since by and large it was only there in the first place as a result of impaired excretion/degradation.”
The Conboys don’t embrace the programmed aging perspective, but neither is their understanding of what they see the same as Aubrey’s. The way Irina explained it to me is that the age of the biological of the body is simply a measure of how much damage has accumulated, but that cycles of epigenetics and catalysis are self-reinforcing.
“Epigenetic, mRNA, and protein are steps of one process, regulation of gene expression. And none of these steps are permanent they all actively and constantly respond to cell environment — tissue and systemic milieu…With aging there is a drift which is re-calibrated by a number of rejuvenation approaches…When an auto-inductive age-elevated ligand is diluted, it cannot activate its own receptor and induce its own mRNA, so ligand levels diminish to their younger states for prolonged time.”
The Conboys theorize that these harmful proteins are part of a positive feedback loop, in other words, a cycle that is self-sustaining
epigenetic state ⇒ gene expression ⇒ translation to circulating proteins ⇒ feedback that alters the epigenetic state
With age, the body has slipped into a dysfunctional, self-sustaining cycle, and with the shock of disruption, they are able to nudge it back into a more robust and youthful cycle, also self-sustaining.
Figure 6. Model of the dilution effect in resetting of circulatory proteome. System: A induces itself (A, red), and C (blue); A represses B (green), C represses A. A dilution of an age-elevated protein (A, at D1: initial dilution event), breaks the autoinduction and diminishes the levels of A (event 1, red arrow); the secondary target of A (B, at event 2 green arrow), then becomes de-repressed and elevated (B induces B is postulated); the attenuator of A (C, at event 3 blue arrow), has a time-delay (TD) of being diminished, as it is intracellular and was not immediately diluted, and some protein levels persist even after the lower induction of C by A. C decreases (no longer induced by A), and a re-boot of A results in the re-induction of C by A (event 4 blue arrow) leading to the secondary decrease of A signaling intensity/autoinduction, and a secondary upward wave of B (events 5 red arrow and 6 green arrow, respectively). alpha = 0.01, kc = 0.01, beta = 0.05, epsilon = 0.1, ka = 0.1. Protein removal rates from system: removalA = 0.01, removalB = 0.1, removalC = 0.01, Initial values: initialA = 1000, initialB = 400, initialC. = 700
For me, the surprising thing in Irina’s account is that there is no hysteresis in this system. The reprogramming responds to changes in the blood levels of signals within minutes. It is difficult for such a system to be homeostatic. I wonder how that can be. Life is all about homeostasis, and intuitively, we all imagine that negative feedback loops are more common than positive feedback loops. (Negative feedback loops lead to homeostasis; positive feedback loops lead to runaway, exponential change.)
Is there a clock somewhere? Is the brain special?
In the Conboy view, signals in the blood are emitted from all over the body, and not especially from the hypothalamus. If brain tissue responds in a seemingly exceptional way to proteins in the blood, it is because of selective passage of those proteins by the blood-brain barrier.
The authors remind us that in past parabiosis experiments (where blood is exchanged between old and young mice), the brain tissue of the young mice grew older but brains of the old mice didn’t get younger. This was an indication that brain aging is caused by affirmative action of “bad actors” in the plasma, and that these are able to penetrate the blood-brain barrier. This observation was part of the inspiration for the current experiments.
The corresponding procedure in humans is already FDA approved
Therapeutic Plasma Exchange (TPE) is a well-established medical procedure, and has already been performed on an experimental basis by co-author Dobri Kiprov. There is anecotal history of suggestive results, which I will write about in my next post.
Comparison with Katcher’s Elixir
This week’s announcement from the Conboys and last month’s preprint from Katcher/Horvath come from the same school of thought: that aging is coordinated through the body by signal molecules in the blood. Both demonstrated dramatic rejuvenation in rodents based on a short-term intervention, and both have plans for commercialization and human trials to begin ASAP.
So it is curious that in other ways, the programs of Katcher and Conboy are so different.
- While both approaches are rooted in differing compositions of blood plasma between young and old, the Conboys focus exclusively on removing species that are inhibiting youthful regeneration, while Katcher’s approach is to add back the proteins that formerly kept the animal young.
- The Conboys have fully disclosed all aspects of their experimental protocol, whereas the content of Katcher’s elixir remains a trade secret.
- Katcher is on the fringe of academic research, and the Conboys’ lab is at one of the premier academic institutions in the world.
- Katcher is a year further along, having experimented with different dosages and timings. Neither Katcher nor the Conboy lab has yet demonstrated life extension.
- The Conboys demonstrate rejuvenation with wound healing, tissue structure, and renewal of nerve growth. Katcher’s claim is based on physiology (especially inflammation), cognitive performance, and methylation clock algorithms.
- In fact, Katcher regards restoration of youthful methylation patterns as the best evidence he could offer for rejuvenation (I agree), while the Conboys are reserving judgment about the importance of methylation, and bristle at the language of a methylation “clock”.
- Katcher understands the effects of plasma transfusions in terms of a broad theory (which I support). Aging is an epigenetic program, governed and enforced by a “clock” that operates via a feedback loop between circulating proteins that govern gene expression and gene expression that generate those proteins. The Conboys recognize they are working this feedback loop (their Fig 6) but they resist the theory that it is the essential cause of aging.
My guess is that a combination of their two approaches will be necessary for full remediation of aging, and that a combination of their resources, credibility, theoretical foundations, and contacts would be a transformative event for medical science, for biotech industry, and for biological theory. It is my fervent hope that Katcher and the Conboys might work together.