Two new papers are changing how scientists think about aging. One, published today in Nature by Vadim Gladyshev and colleagues at Harvard, and a second in Cell by Juan Carlos Izpisua Belmonte and his team at Altos Labs, argue that aging isn’t just wear and tear like rust on a car. It’s a loss of cell identity.
Until recently, the dominant model was that aging happened because cells accumulated damage over time. The new model adds a second pathway: cells lose their sense of who they are. We don’t know yet which model matters more or if they work together. But the evidence is mounting.
Establishing Cellular Identity
Cells start building their identity in the womb. Once that identity is set, it stays fixed for life — except when it falls apart during aging. Every cell in your body carries the same DNA, but the way that DNA is packed and marked is different from cell to cell. That packing — the epigenetic structure — includes methylation, histones, nucleosomes, and chromatin, and it forms a three-dimensional code that defines what each cell does.
The cell’s command center is the nucleus, and the epigenetic structure is the bedrock of that identity. There’s a three-tier system that keeps that identity stable:
- The fast layer responds to acute stress within minutes to hours, using transcription factors like AP-1.
- The intermediate layer takes days to weeks to shift states, turning a cell into a healing mode and back to normal.
- The slow layer is the lock. It holds the identity in place through chromatin architecture.
The Waddington Landscape
Think of cell identity as a valley in a landscape. A cell sitting in that valley represents a specialized role — brain cell, liver cell, kidney cell. The steep slope around the valley keeps the cell from sliding out. With aging, that slope flattens. The valley loses its shape, the basin gets shallow, and the rules that hold the cell in place break down. That’s loss of epigenetic constraints, and it lets the cell drift toward a different state.
The drift moves cells toward a mesenchymal state, which looks like fibroblasts — connective tissue cells that lay down scar tissue. The drift happens across 46 tissue types, and it’s tied to disease progression and poor outcomes, from atherosclerosis to age-related macular degeneration to Alzheimer’s disease.
PRC2 as Sculptor
PRC2 is the sculptor of that landscape. It builds the deep epigenetic constraints that define each cell’s identity. When chronic inflammation overrides the fast layer, it blocks PRC2 and erodes the slow layer. Even after the inflammation clears, the damage persists in chromatin. That loss of epigenetic grammar can become the foundation for cancer cells. Eventually, the slow layer collapses entirely, and the Waddington landscape itself falls apart.
Epigenetic Clocks
Epigenetic clocks have been used to predict lifespan across 348 mammalian species, including humans. But researchers didn’t know what the clocks were actually measuring. Now they do: it’s PRC2, specifically the pace of slow layer erosion. The PRC2 low-methylated regions mark the rate of decay.
Freezing the Drift
The Cell paper shows a positive feedback loop. Fibroblasts activate other fibroblasts, driving more scarring. How do you stop that?
One approach is caloric restriction (CR). CR reduces acetyl CoA, which in turn reduces PRC2 methylation. That preserves the slow layer architecture, conceptually freezing the drift. CR has been shown to extend lifespan in mammalian species, though results in non-human primates have been inconsistent.
What This Changes
The model offers a new target for intervention. Instead of treating the symptoms of aging, the goal becomes preserving the instructions that tell cells what to be. That’s a different strategy entirely.
The papers are still early. The relative contributions of wear and tear versus identity loss aren’t settled. But the model is compelling enough that researchers are already testing interventions.
| Model | Key Mechanism |
|---|---|
| Wear and tear | Accumulated damage |
| Loss of identity | Erosion of epigenetic constraints |
What Happens Next
The field is moving fast. Belmonte’s team has already verified the drift across 46 tissue types. The model is being tested in living systems, and the results are coming in.
The question is whether the drift can be reversed. The feedback loop suggests it may be self-sustaining. But CR shows that the slow layer can be preserved. Whether it can be restored remains open.
For now, the model stands as a new way of thinking about aging. The cell is no longer a machine that breaks. It’s a code that erodes. The question is whether that code can be rewritten.
Source material: “Loss of cell identity drives human aging: Two new papers,” erictopol.substack.com.
Get the Notebook.
The day's best stories and every fresh verdict, in plain English, in your inbox by seven. One email a day, no more.