Mind & Body · Friday, 28 August 2026
Your cells die by the hundred billion every day. What makes it safe is the eating, not the dying.
Somewhere between 100 and 300 billion of your cells die today. You notice nothing, because other cells eat the remains within minutes. Most of the diseases in this research are collection failures, not dying failures.
148bn
cells dying in a day
one review's figure; another says 200 to 300 billion
~1 million
cells dying every second
the same review's other way of putting it
3,000
approved drugs screened
to find one that makes macrophages clear faster
17
people in the senolytic brain trials
two open-label studies, five and twelve, no control group
The lead story — what happened
-
Your cells are dying constantly. One review puts the daily total near 148 billion; another puts it at 200 to 300 billion.
[1] [2] -
That is roughly a million cells a second, and you feel none of it.
[2] -
Dead cells are almost never found sitting in tissue. That absence is how researchers know the removal is fast.
[1] -
The eating has a name: efferocytosis. Macrophages, the immune system's cleaners, do most of it.
[5] [6] -
Clearing debris this way is classed as housekeeping. It happens without any immune response being switched on.
[51] -
A dying cell flags itself first. It flips a fat molecule called phosphatidylserine to its outer surface as an eat-me sign.
[1] [5] -
Healthy cells carry a don't-eat-me sign called CD47, which tells a macrophage to leave them alone.
[5] [7] -
A cell that dies tidily makes no noise. The membrane holds, the contents stay in, and no alarm is raised.
[1] -
A cell that dies messily spills its contents. Those spilled parts are the alarm, and the inflammation follows from them.
[27] [28] [49] -
So the harm is not in the dying. It is in a body that sits there uncollected.
[6] [56] -
In artery plaque, cells switch on the don't-eat-me sign and clearing stops. The dead pile up into the plaque's soft core.
[7] [8] -
In lupus, clearing runs slow, and the leftover cell parts become the material the immune system attacks.
[19] [20] -
Senescent cells make themselves hard to remove. They switch on survival pathways that let them dodge the immune clearing.
[16] [18] -
Clearing also slows with age. It fell in old mice, and the same signal was raised in liver samples from older people.
[11] -
Ageing is described as a decline in the body's own ability to repair itself, and this is one of the ways.
[52] [17]
What is pushing on this
a dying cell turns a fat molecule outward so the cleaners can find it
CD47 shuts clearing down inside artery plaque, and the dead pile up
removal fell in old mice, and the same signal was raised in older human liver
Who is involved
How it unfolded
-
Minute 0
the cell switches on caspases, the enzymes that take it apart from the inside
[1] -
Minutes
it flips phosphatidylserine outward as an eat-me flag, while the membrane still holds
[1] [5] -
Shortly after
it releases find-me signals that pull macrophages toward it
[1] -
On contact
receptors called MerTK, Axl and Tim-4 grip the flag, and the cell is swallowed
[6] [24] -
After the meal
the eater shifts to an anti-inflammatory setting instead of raising an alarm
[1] [5] -
If nobody comes
the membrane fails, the contents spill, and the alarm goes off
[27] [28]
Where this points
Watch whether restoring clearance works in people rather than mice. Nobody has yet shown that blocking the EP2 receptor lifts removal in aged human tissue.
The rest of the day
24 more stories on this beat.
Each with its own sources. None of these is a link to the story above.
-
02
The disposal system ages first
Stanford researchers found old mice had built up senescent neutrophils because tissue macrophages had stopped clearing them, traced to a molecule called prostaglandin E2.
[11] Why it matters — It moves the target from killing the cells to repairing the service that should remove them.
-
03
Senolytic evidence is seventeen people
The two published senolytic trials in Alzheimer's, SToMP-AD and STAMINA, were open-label Phase 1 studies with five and twelve participants and no control group.
[12] Why it matters — The phrase zombie cells has travelled much further than the human data behind it.
-
04
Plaque cells hide behind a stop sign
In advanced artery plaque, cells display CD47, a don't-eat-me signal that blocks macrophages. Antibodies against CD47 restored clearing and shrank the dead core in mice.
[7] [8] Why it matters — The plaque's dangerous soft centre is uncollected bodies, and the same death pathways drive calcium into vessel walls.
[50] -
05
Five patients, one broken clean-up gene
Researchers reported five lupus patients carrying recessive mutations in PLD4, a gene whose product breaks down leftover single-stranded nucleic acid. The enzyme did not work.
[22] Why it matters — It is the closest thing here to a test of whether failed clearing alone can start the disease.
-
06
Lupus runs on uncollected debris
In lupus, dead-cell debris and neutrophil extracellular traps are cleared poorly. That leaves a steady supply of the body's own material for antibodies to attack.
[19] [21] [25] Why it matters — The self being attacked is largely rubbish that should have been taken away.
-
07
Sunlight starts it through dead skin cells
Ultraviolet light kills keratinocytes in the skin. The nucleic acids and danger signals they release switch on the interferon response that drives skin lupus.
[23] Why it matters — It explains why sun is a trigger, and it is a matter for a dermatologist, not a search engine.
-
08
The brain's cleaners prune too hard
Microglia and the complement system tag and remove synapses. In Alzheimer's that tagging goes wrong in some regions, while their clearing of the aggregates weakens.
[29] [31] [32] Why it matters — One cell is doing too much of one job and too little of the other, and viral infections can shift the balance too.
[33] -
09
Synapses vanish before neurons do
In a mouse model of inflammation-driven Parkinson's, midbrain synapses were lost one day after the last injection. The dopamine neurons only died at day 14.
[30] Why it matters — The removal came first, which is an argument about cause rather than proof of it.
-
10
Tumours survive by refusing to die
Many cancers block apoptosis using proteins such as survivin and XIAP. Researchers are now testing other death routes to get around that resistance.
[34] [35] [55] Why it matters — Where the body's problem is too little clearing, a tumour's is a cell that will not leave.
-
11
The death enzyme also feeds growth
Caspase-3, best known for carrying out apoptosis, also pushes neighbouring cells to multiply. In tumours that turns an executioner into a growth signal.
[36] [37] Why it matters — A dying cell is not a neutral event; it changes what its neighbours do next.
-
12
Over twenty ways a cell can die
Apoptosis is only one route. Reviews now list necroptosis, pyroptosis, ferroptosis, cuproptosis and more. One counts nineteen named forms; another says over twenty.
[26] [27] [28] Why it matters — The count is unsettled, and each route leaves a different mess for the cleaners.
-
13
Fasting's clean-up claim outruns the data
A review of 22 studies found intermittent fasting produced about 3% weight loss, below the 5% doctors call clinically meaningful. Its lead author called it no miracle solution.
[38] Why it matters — Autophagy, the clean-up inside one cell, is the mechanism most often named for fasting, and the human evidence stays inconsistent.
[41] [39] -
14
Seven days, forty-six healthy adults
A randomised study put 46 healthy adults on 850 calories a day for a week, comparing a low-protein and a high-protein version against a normal diet.
[40] Why it matters — That is the size of the human evidence sitting under a very large diet industry.
-
15
Zombie cells scar the liver
Mayo Clinic researchers found that RNA leaking out of the mitochondria of senescent cells triggers the inflammation driving a severe form of fatty liver disease.
[13] Why it matters — It names a specific route from one uncleared cell to organ damage.
-
16
A new tag for hidden zombie cells
Mayo used aptamers, short synthetic DNA strands that fold into shapes, sifted from over 100 trillion random sequences, to flag senescent mouse cells among healthy ones.
[15] [14] Why it matters — Nobody can count what they cannot see, which is why the field argues about how many there are.
-
17
The cells you cannot replace set the ceiling
A model estimated that damage to neurons and heart muscle cuts a theoretical 1,759-year lifespan to 156 years. Liver, which replaces its cells, lasted millennia.
[42] Why it matters — Turnover is not decay. The tissues that turn over are the ones that last.
-
18
Cooling only partly stops the messy route
After oxygen loss at birth, necroptosis, an inflammatory kind of cell death, starts early and stays active for days. Cooling the baby suppresses it only in part.
[43] Why it matters — The standard treatment falls short against the death route that spills rather than the one that packages.
-
19
Dying cells leave a marked spot
Researchers found that a cell pulling itself apart leaves an anchored patch of membrane behind. The patch carries the eat-me flag and marks the exact site for cleaners.
[9] Why it matters — Under a virus it can also hold virus particles, so the marker becomes a route of infection.
[9] -
20
Diabetic wounds stay open because clearing stalls
In chronic wounds linked to diabetes, macrophage clearing of dead cells is impaired. Inflammation never resolves, and the wound does not close.
[10] [3] Why it matters — The visible failure is a wound; the underlying failure is a collection service running behind.
-
21
Muscle uses three disposal routes at once
In skeletal muscle, apoptosis removes damaged nuclei, necroptosis holds back excessive inflammation, and autophagy breaks down faulty proteins and worn-out parts.
[44] Why it matters — One tissue running three clean-up systems shows how much of staying well is removal work.
-
22
The clean-up inside a single cell
Cells run their own disposal network of chaperones, a protein shredder and the lysosome. Its collapse is now treated as a shared feature of several diseases.
[46] Why it matters — A Tau mutation that causes dementia was found to break the very route that would have cleared Tau.
[47] -
23
Motor neurons wear out fastest
Watching disposal cell by cell in the zebrafish spinal cord, researchers found motor neurons degrade their own contents fastest, and losing a protein called TDP-43 made it worse.
[48] Why it matters — It offers a reason why one cell type dies first in motor neurone disease.
-
24
Bone cleaners change what bone does next
Macrophages that ate dying bone-building cells switched to burning sugar and poured out lactate, which altered inflammation and the making of bone-dissolving cells.
[45] Why it matters — Clearing is not neutral tidying; the act of eating changes the eater and its neighbourhood.
-
25
One compound aimed at two death routes
Researchers describe a molecule, Zharp1-163, that blocks both ferroptosis and necroptosis at once, aimed at inflammatory disease and sudden kidney injury.
[54] [53] Why it matters — Where medicine once tried to stop cells dying, it is now trying to choose how they die.
The trouble is never the ending. It is what gets left lying there.
Endings happen constantly and cost nothing while something clears up after them, so failure shows up first in the clearing, not the ending.
The twist
Nothing goes wrong when a thing ends. It goes wrong when the ending is still lying there tomorrow, and the clearing is the part nobody was watching.
How it works
- Things end constantly, on purpose
- A second system carries the remains away
- Done fast enough, an ending leaves no trace
- That removal is invisible, so nobody counts it
- When removal slows, the remains stay put
- The pile, not the ending, does the damage
Where you've seen this
Demolition
a building taken down to a plan harms nobody; one left half-collapsed does
A kitchen
cooking the meal is not the work that decides the evening; keeping up with the plates is
Switching off old software
turning the system off is the easy day; the data it leaves behind is the expensive decade
The catch
Faster clearing is not always better. The same cleaners trim connections in the brain, and in Alzheimer's the trouble looks like too much removal, not too little.
And the whole of it
Almost everything keeping you well right now is a removal you will never feel. Only the endings show; the clearing stays invisible until it stops.
The Backlog
Spend ten days on either fewer deaths or faster clearing, and find where cutting the dying stops helping.
What is really going on
The dying is normal and constant. Most of the harm described in this research comes from remains that were not collected in time.
Why it works on us — The phrase zombie cells turns a housekeeping failure into an invasion. An invasion has an obvious answer, kill them, and that is the answer being funded. [11][12]
Who gains
-
Senolytic developers and the longevity market
— The zombie-cell framing sells removal as the fix while the published human trials remain five and twelve people.
[12] [14] -
Cancer drug programmes
— Reframing chemotherapy failure as resistance to apoptosis opens a whole new set of death pathways to target.
[34] [55] -
The intermittent-fasting industry
— Cellular clean-up supplies a mechanism story for a diet whose measured weight effect is about 3%.
[38] [41] -
Groups naming new death pathways
— Each named route brings its own literature and its own candidate drugs; one compound already targets two at once.
[26] [54]
Who pays
-
People with lupus
— Their disease runs partly on a clearing failure, and the leftover material becomes what their own antibodies attack.
[19] [21] [22] -
People with diabetes and chronic wounds
— Macrophage clearing is impaired, so inflammation does not resolve and the wound stays open.
[10] [3] -
Older adults
— Clearing capacity falls with age, and what is marketed to them is killing cells rather than restoring the service.
[11] [17] -
Newborns after oxygen loss at birth
— The inflammatory death route runs on for days, and cooling the baby suppresses it only in part.
[43]
What nobody knows yet
Open questions from across today’s stories — ours included.
-
01
How many of your cells actually die in a day.
Four reviews give four figures: nearly 100 billion
[4] , about 148 billion[1] , about 0.4% of 37.2 trillion[3] , and 200 to 300 billion[2] . -
02
How many distinct ways a cell can die.
One review counts nineteen named forms of regulated cell death; another says over twenty. The list is still being written.
[27] [28] -
03
Whether restoring clearance would help people, not just mice.
The Stanford result is in mice. The human liver data is correlative, and the team says the receptor block has not been tested in aged human tissue.
[11] -
04
Whether senolytics do anything for Alzheimer's.
The published human evidence is two open-label Phase 1 trials with five and twelve people and no control arm. A Phase 2 trial is under way.
[12] -
05
Whether fasting works through cellular clean-up, or works much at all.
Researchers say the hardest problem is that nobody knows what people in these studies actually ate.
[39] [38] -
06
Whether too much pruning causes brain disease or follows it.
Synapse loss came 13 days before neuron death in one mouse model, but that model is systemic inflammation, not human Parkinson's.
[30] -
07
Whether switching off a don't-eat-me signal is safe in people.
Several clinical programmes aimed at boosting clearance were stopped early, which is why the search moved to already-approved drugs.
[2]
Most of what keeps you well is work you will never feel. It has been going on quietly, without a break, for every second you have been alive.
More from Mind & Body