Stem Cells
The Repair Crew.
And Why Yours Is Shrinking.
You replace around 330 billion cells a day. Something has to make them. That something is a small, quiet population of stem cells sitting in your bone marrow, your gut lining, your muscle, your skin. Their number does not simply fall with age. Their diversity collapses. And in the female body, the hormone that maintains several of these populations withdraws two decades before anyone starts calling it ageing.
I was at a health conference a while back, and somewhere in the programme there was a talk about stem cells. Lovely talk. I went in curious and came out fascinated, which does not happen as often as I would like at conferences.
I already knew stem cells from my own journey. Not from a textbook. From fasting. I once did a longer fast, past 72 hours, and for the whole stretch of it my back pain was close to paralysing. I could not do much of anything. I lay there and waited it out.
And then afterwards it was like I had received an entire new back.
I have had the longest trouble with my back. It is still not fully healed, because that is close to impossible, and that is a story for another time. But this thing. This one thing. Amazing.
So that is where this post came from. A talk, a fast, and a back that behaved differently for a while. What I want to give you here is the part I went looking for afterwards: what stem cells actually are, what they are responsible for in your body, how fast they decline while you age (the honest answer is unsettling, though not in the way most wellness content tells you), and what the evidence really says about how to multiply them. Naturally, and chemically.
Some of what follows is beautiful. Some of it is being sold to you at four thousand euros a session on evidence that does not exist yet. I will be clear about which is which.
Your Repair Crew Across a Lifetime
Drag the age. Watch what actually changes, and when.
Curve is schematic. It renders the pattern described by Mitchell and colleagues at the Wellcome Sanger Institute, who sequenced blood stem cell clones across ten people aged 0 to 81. It is not plotted from individual participant data, and no single person's curve looks exactly like this. [2]
A stem cell is defined by two things it can do that no other cell can do at once. It can make a copy of itself, which is called self-renewal. And it can produce a specialised daughter cell that becomes something else: a red blood cell, a gut lining cell, a muscle fibre, a bone cell. Self-renewal plus differentiation. That combination is the whole definition, and everything sold under the phrase "stem cell" either meets it or does not.
Most of them are not floating around your bloodstream waiting to be summoned. They sit in specific physical locations called niches: the haematopoietic stem cells in your bone marrow, the satellite cells wedged between a muscle fibre and its surrounding membrane, the intestinal stem cells at the base of each crypt in your gut lining, the mesenchymal cells in marrow and fat. The niche is not decoration. It holds them dormant, and dormancy is what protects them. A stem cell that divides constantly accumulates mutations and burns out. A stem cell that stays quiet lasts decades.
The scale of what they are responsible for is easy to underestimate. Ron Sender and Ron Milo at the Weizmann Institute built the first full census of cellular turnover in the human body. Roughly 330 billion cells are replaced every day. Around 80 grams of cell mass, daily. Close to 90 percent of that turnover is blood. Your intestinal lining is rebuilt every three to five days. Neutrophils last under a week. Red blood cells last about 120 days. None of those cells divide to replace themselves. A stem cell upstream makes every one of them. [1]
Which means stem cell decline is not an abstraction about ageing. It is the supply line for a tissue that rebuilds itself fifty to a hundred times over a lifetime. When the supply line narrows, the tissue does not fail on a particular Tuesday. It just recovers a little worse, each time, for years.
Here is where most of what you have read about stem cells and ageing is wrong, or at least imprecise. You will see a number circulating everywhere: that you have one mesenchymal stem cell per 10,000 bone marrow cells as a newborn, and one per 2,000,000 by age 80. It is a striking number. It traces back to estimates published by Arnold Caplan at Case Western Reserve in 2007, not to a study that followed anyone across a lifetime. [10] Treat it as an order-of-magnitude impression rather than a measurement.
The measurement that does exist is better, and stranger.
Emily Mitchell and colleagues at the Wellcome Sanger Institute sequenced the genomes of individual blood stem cell clones in people aged 0 to 81 and reconstructed their family trees. Before 65, blood production is massively polyclonal: somewhere between 20,000 and 200,000 stem and progenitor cells contributing evenly, with at most one expanded clone per person and that clone contributing under 2 percent of the total. After 75, between 12 and 18 independent clones account for 30 to 60 percent of all blood production, some single clones carrying a third of it. Clonal diversity does not taper. It drops off a cliff at around age 70. [2]
The mutations were not new. Each stem cell picks up roughly 17 mutations per year, steadily, for your whole life. Most of the clones that take over after 70 arose decades earlier, between birth and age 40. They sat there for forty years, expanding slowly, and then the balance tipped. Only 22 percent carried a known cancer driver gene, which means the selection pressure is broader than anyone had mapped. [2]
So the honest answer to "how fast are they declining" is not a smooth percentage per decade. Count goes down. Function goes down. But the thing that changes most sharply is variety, and variety is what lets a tissue respond to something it has not seen before. A blood system run by fifteen clones is a blood system with fifteen answers to every question.
Your repair crew does not resign. It consolidates. A handful of clones take the work, the rest go quiet, and the redundancy that made you resilient at thirty is simply no longer there at seventy-five.
The Fasting Clock
What the research has actually measured, hour by hour. Drag the dial. Read the evidence badge before you believe anything.
Every panel names the species the finding comes from. Where a claim rests on mice, it says so. This is a question where most of the internet does not bother.
The word "multiply" needs a caveat before we go further. Almost nothing in the natural-intervention literature increases your total stem cell count in the way a supplement ad implies. What these interventions do is change the state of the cells you already have: waking dormant ones, restoring their capacity to activate, shifting their metabolism, or mobilising them out of the niche and into circulation. That is a real effect. It is not the same claim.
Prolonged fasting shuts down the signal that keeps blood stem cells dormant. Chia-Wei Cheng and Valter Longo at the University of Southern California ran cycles of prolonged fasting in mice and found the mechanism: fasting lowered circulating IGF-1, IGF-1 signalling lowered PKA activity inside haematopoietic stem cells, and the drop in PKA flipped those cells from dormancy into self-renewal. White blood cell counts fell during the fast and rebuilt on refeeding, with the damaged cells cleared out in the process. They also ran a small trial in people undergoing chemotherapy who fasted for 72 hours beforehand, where IGF-1 fell and the fasted patients tolerated treatment better. The mechanism is mouse work. The human arm is a small early-phase trial in cancer patients, not a longevity protocol. [3]
A shorter fast changes what intestinal stem cells run on. Maria Mihaylova and David Sabatini at the Whitehead Institute fasted mice for 24 hours and found their intestinal stem cells had switched from glucose to fatty acid oxidation through the PPAR transcription factors. Regenerative capacity in culture doubled, and this held in aged mice as strongly as in young ones. When they blocked the PPAR pathway, the benefit disappeared. They also reproduced part of the effect with a drug that mimics PPAR activation, which matters for anyone who cannot safely fast. [4]
Exercise restores the ability of old muscle stem cells to wake up. Jamie Brett and Thomas Rando at Stanford gave aged mice access to a running wheel. The muscle stem cells of the old runners regained their activation capacity, and muscle repaired faster. The mechanism was the restoration of Cyclin D1, which had declined with age, acting by suppressing TGF-beta signalling. The effect appeared only in old animals. In young mice, whose stem cells already activated normally, running changed nothing. [5]
Sleep loss ages the blood stem cell pool directly, and part of that persists. Cameron McAlpine and Filip Swirski at Mount Sinai fragmented sleep in mice and restricted sleep in healthy human volunteers to around six hours a night for six weeks. In both, stem and progenitor cells proliferated more and monocytes rose. In the mice, the increased turnover homogenised the pool and accelerated the loss of clonal diversity, the same measure that collapses with age. After ten weeks of recovery sleep, most markers normalised. Thirty-one percent of the epigenetic changes did not. [6]
Read those four together and a pattern shows up. Fasting, exercise and sleep are not adding stem cells. They are protecting dormancy, restoring activation capacity, and slowing the drift toward a smaller number of dominant clones. The lever is not quantity. It is preserving the state that keeps quantity useful.
There is a pharmacology that genuinely multiplies circulating stem cells, and almost nobody selling stem cell treatments is using it. Granulocyte colony-stimulating factor, given as filgrastim, drives haematopoietic stem cells out of the marrow niche and into peripheral blood over several days. Plerixafor blocks the CXCR4 receptor that tethers them to the niche in the first place, and gets added when G-CSF alone does not yield enough. This is how a bone marrow donation is collected from an arm vein instead of a hip bone. It works, it is approved, and it exists to harvest cells for transplant. Nobody prescribes it to feel younger, because mobilising your stem cells into your bloodstream is not the same as regenerating anything.
Which brings us to what is actually for sale. The United States Food and Drug Administration is direct about it: the only approved stem cell products are blood-forming stem cells derived from umbilical cord blood, used for disorders of blood production. There are no approved exosome products. Everything offered for arthritis, back pain, hair, skin, autoimmune disease, neurological conditions or general rejuvenation sits outside that. [8]
And there is a naming problem underneath the market. In 2017 Arnold Caplan, the researcher who coined the term "mesenchymal stem cell" in the 1990s, published a paper asking the field to stop using it. His argument: the cells being injected into knees and faces around the world are, in the vast majority of cases, not functioning as stem cells at all. They are secreting signalling molecules that modulate the local immune environment. He proposed renaming them medicinal signalling cells, keeping the initials so the literature would not break. [9]
The Evidence Ladder
Twelve things marketed as ways to increase your stem cells. Filter by what the evidence actually is. Tap any card for the detail.
Estrogen is a stem cell hormone, and menopause is a stem cell event nobody calls one
Daisuke Nakada and Sean Morrison, then at UT Southwestern, found that female mice divide their haematopoietic stem cells more than males do, and that the difference comes from the ovaries rather than the testes. Blood stem cells carry estrogen receptor alpha. Give estradiol, and division increases in both sexes. During pregnancy, estrogen drives stem cell expansion and spleen erythropoiesis to meet the demand. Delete the receptor from blood cells, and the female advantage and the pregnancy expansion both vanish. [7]
Brittany Collins and Dawn Lowe at the University of Minnesota found the same logic in muscle. Remove the ovaries in mice and satellite cell numbers fall by 30 to 60 percent, through apoptosis, with a 3.7-fold increase in dying stem cells. Strength recovery after injury dropped 19 percent, and after a second injury 23 to 39 percent. Then they did the study that should have changed clinical conversation: they biopsied women through the menopause transition, longitudinally. Satellite cells fell about 15 percent from perimenopause to postmenopause within a year, and the drop tracked each woman's falling estradiol. Estrogen replacement and a SERM restored satellite cell numbers in the deficient mice. [11]
Sit with the sequencing. Regenerative medicine clinics will sell a woman in her fifties an injection of cells into her knee for several thousand euros, on evidence the FDA has not approved. Meanwhile the single largest documented hit to her muscle stem cell pool is hormonal, happens on a predictable timeline, was measured in actual women, and gets discussed in clinic as hot flushes and mood. Her repair capacity is being reclassified as a quality-of-life complaint.
This is not anecdote. This is data. And it is data that took until 2014 and 2019 to be published, in a field that has been studying stem cells since the 1960s.
Two honest limits, including one about my own back
First, my fast. Three days without food improved my back for a while, and I felt it clearly enough that I went and read about stem cells afterwards. That does not mean stem cells did it. Seventy-two hours of fasting also lowers systemic inflammation, shifts fluid balance, drops circulating IGF-1, and, in my case, meant three days of not loading a back that does not like being loaded. Any of those could explain what I felt. An n of one with four plausible mechanisms is not evidence for the mechanism you find most beautiful. I am telling you what happened to me. I am not telling you why.
Second, and this is the one that matters for you: the fasting and stem cell literature was not built on women. Sofia Cienfuegos and Krista Varady at the University of Illinois Chicago reviewed every human trial measuring reproductive hormones during intermittent fasting. In premenopausal women with obesity, androgens fell and sex hormone binding globulin rose. Estrogen, FSH, LH and prolactin showed no change across the small number of studies available. Their own summary of the evidence base is unusually blunt: very few studies, samples of 16 to 107, hormones measured as secondary outcomes, and it is highly likely none of the trials were adequately powered. Cycle phase was not standardised. And no trial has been run in perimenopausal or postmenopausal women at all. [12]
So if you are cycling, or underweight, or have a history of restriction, or are pregnant or trying to be, or are managing a condition that affects your energy availability, the honest position is that nobody has studied what a 72-hour fast does to you. Not that it is dangerous. That it is unstudied, which is a different and in some ways more annoying thing. This post explains mechanisms. It does not prescribe, and prolonged fasting is a conversation to have with a clinician who knows your history.
Of everything here, chronic short sleep is the intervention with human data showing it pushes the blood stem cell pool toward the aged pattern, and the only one where a third of the epigenetic changes survived ten weeks of recovery. Recovery sleep repairs most of it. Not all of it. If you are choosing one lever, this is the one where the cost of ignoring it compounds.
Exercise restored activation capacity in old muscle stem cells and did nothing in young ones, which is exactly the profile you want from an ageing intervention. For a woman moving through the menopause transition, this sits alongside a documented 15 percent drop in the satellite cell pool over a single year. Loading muscle is not a beauty activity in that window. It is maintenance on the repair system itself.
The IGF-1 to PKA to self-renewal chain is elegant and reproducible in mice. The human evidence is a small trial in chemotherapy patients. If you fast, do it knowing you are acting ahead of the data rather than on top of it, and knowing the regeneration in those studies happened during refeeding rather than during the fast itself. What you eat afterwards is not an afterthought to the protocol. In the mouse work it is the protocol.
Is this product approved by a regulator for this indication, or offered under an exemption? Are you injecting cells, or a cell-free exosome preparation the FDA has approved none of? Do you claim these cells engraft and become new tissue, or that they signal locally and clear? And where is the randomised trial in people with my condition, at my age, with my hormonal status? A clinic that answers all four in plain language is a different proposition from one that shows you testimonials.
I went into that talk knowing one thing about stem cells, which was that something in my body had behaved unusually after three days without food. I came out with a different question, and it is the one I would leave you with. Not how do I get more of them. How do I stop spending the ones I have.
Because the picture in the research is not a slow leak that you top up. It is a pool that stays wide for six decades and then narrows sharply, while every night of fragmented sleep pushes a little more turnover through it, and while, in the female body, the hormone that has been maintaining several of those populations withdraws somewhere around fifty and takes part of the muscle repair system with it. Nobody framed it to me that way. Nobody frames it that way in a menopause consultation either.
My back is not healed. It probably will not be, and I have made my peace with that being a longer story. But I understand now what those three days may have touched, and, just as usefully, what they may not have. Both of those are worth more to me than the version of this I could have written from a conference talk and a good feeling.
If you want the longer version of this, with the hormonal architecture underneath it laid out properly rather than in one section of a blog post, it is in The Art of Female Health. The chapters on cellular repair and the perimenopausal transition go considerably deeper than I can go here. nammu.academy/book
References
- Sender, R., & Milo, R. (2021). The distribution of cellular turnover in the human body. Nature Medicine, 27(1), 45–48. https://doi.org/10.1038/s41591-020-01182-9
- Mitchell, E., Spencer Chapman, M., Williams, N., et al. (2022). Clonal dynamics of haematopoiesis across the human lifespan. Nature, 606(7913), 343–350. https://doi.org/10.1038/s41586-022-04786-y
- Cheng, C.-W., Adams, G. B., Perin, L., et al. (2014). Prolonged fasting reduces IGF-1/PKA to promote hematopoietic-stem-cell-based regeneration and reverse immunosuppression. Cell Stem Cell, 14(6), 810–823. https://doi.org/10.1016/j.stem.2014.04.014
- Mihaylova, M. M., Cheng, C.-W., Cao, A. Q., et al. (2018). Fasting activates fatty acid oxidation to enhance intestinal stem cell function during homeostasis and aging. Cell Stem Cell, 22(5), 769–778.e4. https://doi.org/10.1016/j.stem.2018.04.001
- Brett, J. O., Arjona, M., Ikeda, M., et al. (2020). Exercise rejuvenates quiescent skeletal muscle stem cells in old mice through restoration of Cyclin D1. Nature Metabolism, 2(4), 307–317. https://doi.org/10.1038/s42255-020-0190-0
- McAlpine, C. S., Kiss, M. G., Zuraikat, F. M., et al. (2022). Sleep exerts lasting effects on hematopoietic stem cell function and diversity. Journal of Experimental Medicine, 219(11), e20220081. https://doi.org/10.1084/jem.20220081
- Nakada, D., Oguro, H., Levi, B. P., et al. (2014). Oestrogen increases haematopoietic stem-cell self-renewal in females and during pregnancy. Nature, 505(7484), 555–558. https://doi.org/10.1038/nature12932
- U.S. Food and Drug Administration. (n.d.). Consumer alert on regenerative medicine products including stem cells and exosomes. https://www.fda.gov/vaccines-blood-biologics/consumers-biologics/consumer-alert-regenerative-medicine-products-including-stem-cells-and-exosomes
- Caplan, A. I. (2017). Mesenchymal stem cells: Time to change the name! Stem Cells Translational Medicine, 6(6), 1445–1451. https://doi.org/10.1002/sctm.17-0051
- Caplan, A. I. (2007). Adult mesenchymal stem cells for tissue engineering versus regenerative medicine. Journal of Cellular Physiology, 213(2), 341–347. https://doi.org/10.1002/jcp.21200
- Collins, B. C., Arpke, R. W., Larson, A. A., et al. (2019). Estrogen regulates the satellite cell compartment in females. Cell Reports, 28(2), 368–381.e6. https://doi.org/10.1016/j.celrep.2019.06.025
- Cienfuegos, S., Corapi, S., Gabel, K., et al. (2022). Effect of intermittent fasting on reproductive hormone levels in females and males: A review of human trials. Nutrients, 14(11), 2343. https://doi.org/10.3390/nu14112343