Knowing the healthcare headlines is easy.

Understanding what they mean for the business? That’s the hard part.

Healthcare Brew is a free newsletter breaking down the forces shaping the healthcare industry—from pharmaceutical developments and health startups to policy shifts, regulation, and tech changing how hospitals and providers operate.

No clinical deep dives. No overstuffed jargon. No guessing what actually matters. Just clear, focused coverage built for the people making decisions behind the scenes.

Join 135K+ administrators and healthcare professionals staying informed, for free.

Check it out

New to LEVITY? Start here! Want to know more about who’s behind LEVITY? Check out this page. 🙏🏼 Not subscribed to the LEVITY podcast on Youtube yet? Do it here. 🎧 More of a listener? The podcast is also available on Spotify, Apple Podcasts and other places.

There is a question that arises once you start talking about reprogramming neurons. If you could wind back the biological clock of a neuron - strip away the accumulated epigenetic damage, restore its youthful gene expression - would you also erase the information it carries? Would you lose the memory encoded in that cell?

This is not an abstract philosophical worry. It is an engineering challenge for anyone who wants to rejuvenate the brain. Unlike a liver or a patch of skin, the brain is not a generic tissue. It stores who you are. Every approach to neuronal reprogramming must thread a needle: make the cell younger without making it forget.

Two recent studies* now suggest this needle can be threaded - and they do so from complementary directions that together paint a more complete picture than either offers alone.

* Several other recent rodent studies, using less memory-targeted designs, point in the same direction: partial reprogramming can improve cognition and/or blunt neurodegeneration signals in the brain.

The headline paper, published in Neuron recently, by Gabriel Berdugo-Vega and colleagues in Johannes Gräff's lab at EPFL (École Polytechnique Fédérale de Lausanne), takes what might be the most elegant approach to brain reprogramming yet attempted. Rather than targeting neurons broadly, the team went after engram cells - the sparse, specific groups of neurons that are activated during learning and reactivated during recall. These are, in a very literal sense, the cells where particular memories live.

The setup was clever. The researchers designed a genetic system that could do two things at once: first, flag whichever neurons fired during a learning event, and then deliver a brief pulse of three Yamanaka factors - Oct4, Sox2, and Klf4, collectively called OSK - to those flagged cells and only those cells. The pulse was short-lived and the targeting was precise: not “neurons in the hippocampus” but specifically the neuronal ensemble that encoded a given memory.

They tested this in two brain regions. In the dentate gyrus of the hippocampus, where recent memories form, engram reprogramming restored fear memory in aged mice to levels seen in young animals. In the medial prefrontal cortex, which supports remote memory consolidation, the same approach rescued recall of memories formed weeks earlier. Critically, the reprogrammed engram neurons were preferentially reactivated during recall compared to their non-reprogrammed neighbors, suggesting the treatment actually strengthened the memory trace rather than blurring it.

The team then moved to Alzheimer's disease models. In APP/PS1 mice, OSK-reprogrammed engrams in the dentate gyrus shifted the animals' navigation strategies back toward hippocampus-dependent spatial learning - a capacity that typically degrades as amyloid pathology accumulates. In a more aggressive Alzheimer's mouse model called 5xFAD, which carries five mutations linked to familial Alzheimer's and develops pathology faster, reprogramming dentate gyrus engrams was used to validate the approach further and feed into the team's cognitive clock analysis.

What made this work at a molecular level? The team used a technique that reads both gene expression and chromatin accessibility from the same individual cells, giving them an unusually detailed picture of what OSK actually changed. They found that aging and Alzheimer's pathology had shut down access to specific stretches of DNA in engram neurons - stretches that control genes involved in synaptic function.

One finding stood out: a potassium channel gene called Kcnj3, which helps regulate how easily a neuron fires, had gone quiet in aged and diseased cells. In the regions of DNA that control this gene, the team found an enrichment of Klf4 binding motifs - suggesting that Klf4, one of the three OSK factors, was likely involved in reopening access to these regulatory stretches and restoring the channel's expression. The result was that hyperexcitable neurons (a hallmark of both aging and Alzheimer's) calmed back down to normal firing patterns.

The team also built what they call a “cognitive clock”, a statistical model that uses water-maze learning performance to estimate an animal's age. OSK-treated aged and AD mice scored younger on this clock, meaning their learning behavior had shifted back toward that of younger animals. And the neurons that drove this improvement maintained their identity markers, the molecular signatures that make a dentate gyrus neuron a dentate gyrus neuron. Those markers were not just preserved by the treatment. They were strengthened.

Across multiple models, the signal is that partial reprogramming can restore youthful gene regulation in neurons while preserving neuronal identity - and in engram-targeted experiments, the memory trace isn’t erased; it’s functionally strengthened.

The Gräff lab's work did not emerge in a vacuum. Back in July 2024, Alejandro Ocampo and collaborators - including Yuri Deigin of YouthBio Therapeutics - had posted a preprint taking a less targeted but more translationally immediate approach. Instead of labeling individual engram cells, they used genetic and viral tools to reprogram excitatory neurons broadly within the dentate gyrus, delivering all four Yamanaka factors (OSKM, including c-Myc) rather than just three.

The results followed a pattern that should interest anyone thinking about clinical timelines. In young mice, the treatment produced no detectable behavioral change. Which is to be expected, since you cannot rejuvenate what is not yet aged. In middle-aged mice (six months), cyclic reprogramming improved learning after three to four weekly treatment cycles. In old mice (17–26 months), chronic weekly induction of OSKM starting around 17 months improved performance on Y-maze and contextual fear conditioning tasks. A single injection alone was not enough, the benefits required sustained, repeated cycles.

At the cellular level, the Ocampo/Deigin study showed that reprogrammed neurons maintained their dentate gyrus identity, and that the treatment's behavioral effects were specific to aged animals. The detailed molecular picture of what rejuvenation looks like inside these neurons - the chromatin changes, the rescued potassium channels, the strengthened identity markers - comes from the Gräff lab's engram-specific work described above. The two studies converge on the same conclusion from different angles: reprogramming makes neurons younger without erasing what they are.

YouthBio Therapeutics, the company Deigin co-founded, is now building on the broader approach. In September 2025, the company announced positive feedback from the FDA in a so-called INTERACT meeting, an early-stage consultation where the agency gives guidance before a company formally applies to begin clinical trials. The FDA agreed that existing preclinical data support the bioactivity of YouthBio's lead candidate YB002, a gene therapy designed to transiently express Yamanaka factors in the brain of Alzheimer's patients. It endorsed the company's proposed path toward a first-in-human trial. YouthBio is now focused on manufacturing and toxicology work, with a target of reaching first-in-human dosing in approximately three years.

“What’s exciting is the shift from ‘is this biologically plausible?’ to ‘can we engineer control, specificity, and dosing to make it clinically practical?’ The FDA INTERACT feedback helps focus the next steps on execution - CMC and tox* - rather than speculation”, Deigin tells LEVITY.

* CMC (Chemistry, Manufacturing, and Controls) is the work to prove you can manufacture and quality-control the therapy consistently, and tox is the preclinical safety testing to identify harmful effects and safe dose ranges before human trials.

Deigin has argued that neurons provide a particularly wide safety margin for reprogramming - multiple studies now show they resist dedifferentiation even under prolonged factor expression, unlike some peripheral tissues where the margin is narrower and the risks of aberrant proliferation more real. The brain, ironically, may be among the safest organs to reprogram.

We are still in mice, and the distance from a mouse hippocampus to a human Alzheimer's patient is vast. But the principle has been established in a way it had not been before: the memories are in the structure of the connections and the patterns of activity, not in the age of the cell. You can turn back the clock without rewinding the tape.