How a Pig Virus Accidentally Solved One of Alzheimer's Biggest Puzzles

Memory is protein. That sounds reductive, but it is one of the most solid results in modern neuroscience. When you learn something new — when a memory consolidates from a fragile short-term impression into a durable long-term record — your neurons need to build new proteins. Synapses physically remodel. Connections strengthen. The molecular machinery of the cell has to run at full speed to inscribe that memory into the structure of the brain.

So what happens when that machinery is silenced?

A new study published in Science on April 2, 2026, by researchers at Altos Labs in Redwood City, California, answers that question with uncomfortable clarity — and offers an unexpected solution borrowed from one of nature’s more devastating pathogens.¹

The Brake the Brain Cannot Release

Inside every cell sits a set of molecular alarm systems. When things go wrong — a viral infection, a buildup of misfolded proteins, a shortage of nutrients — these alarms converge on a single molecular switch: the phosphorylation of a protein called eIF2α. This switch does something profoundly useful in a crisis. It tells the ribosome — the cell’s protein-building factory — to slow down. Stop translating most messenger RNAs. Conserve energy. Deal with the emergency.

The whole network is called the integrated stress response, or ISR, and it is an ancient and elegant solution to the problem of cellular survival.

The trouble is what happens when the alarm stays stuck in the on position.

In aging brains, in Alzheimer’s disease, in Down syndrome, and in certain genetic disorders, the ISR activates and does not fully resolve. The cell’s emergency brake stays partially engaged. Protein synthesis is chronically suppressed. New proteins cannot be built. Synapses cannot be remodeled. And memories — the biological events that depend on all of this protein production — cannot form properly.

The cognitive impairments in these conditions are not simply the result of damaged neurons. They are, at least in part, the result of neurons that are still alive but quietly unable to do their job.

A Decade of Clues

This story has been building for more than twenty years. In 2007 and again in 2013, Mauro Costa-Mattioli’s lab — then at Baylor College of Medicine — published landmark papers showing that eIF2α phosphorylation is a master regulator of long-term memory formation.² Mice with genetic modifications that keep eIF2α active (unable to be phosphorylated) had dramatically enhanced long-term memory. Mice with chronically phosphorylated eIF2α — stuck in the stress response — had poor long-term memory.

The implication was startling: the ISR is not just an emergency system. It is a dial that the brain uses to gate the conversion of experiences into lasting memory. When the dial is stuck in the wrong position, the gate stays closed.

In 2015, researchers in the lab of Peter Walter at UCSF discovered a small molecule called ISRIB — for Integrated Stress Response InhiBitor — that could reset this dial. ISRIB works by stabilizing eIF2B, the protein that gets shut down when eIF2α is phosphorylated, allowing translation to continue even under ISR activation.³ In mouse models, ISRIB enhanced spatial memory, reversed age-related cognitive decline, and improved recovery after brain injury. It was a revelation — proof that the ISR could be therapeutically targeted.

But ISRIB was not perfect. It was a small molecule with complex pharmacology, and its effects, while real, were incomplete. The question became: is there something better?

The Virus That Evolved an Answer

Here is where the story takes a remarkable turn.

African Swine Fever Virus — ASFV — is one of the most feared animal pathogens in the world. It causes a catastrophic hemorrhagic fever in domestic pigs and wild boar, with mortality rates approaching 100% in some strains. Outbreaks have decimated pig populations across Africa, Asia, and Europe, and there is currently no effective commercial vaccine. ASFV is a large, complex DNA virus — one of the biggest that infects animals — and one of its strategies for survival is to evade the host cell’s stress response.

When a cell detects a virus, one of its first reactions is to activate the ISR: slow down protein synthesis, make the environment hostile to viral replication. ASFV has evolved a countermeasure. It encodes a protein called DP71L — a short, unusual protein that functions as a powerful inhibitor of the ISR.

DP71L is, structurally, a mimic of a human protein called PPP1R15B (also known as GADD34). Both PPP1R15B and DP71L work by recruiting the phosphatase PP1 to dephosphorylate eIF2α — essentially, turning off the phosphorylation switch that activates the ISR. The virus has, over millions of years, evolved its own version of this human protein, and through ruthless evolutionary pressure, it has become extraordinarily good at the job.

The New Study

The research team led by Lucas Reineke, Ping Jun Zhu, and Mauro Costa-Mattioli — who moved to Altos Labs, the Bezos-backed cellular rejuvenation research institute — set out to understand what makes DP71L so potent.¹

They started by studying PPP1R15B more carefully. A rare human genetic variant called PPP1R15B R658C causes intellectual disability, and the team had not previously understood why. They created mice carrying this variant and found that the mutation destabilizes the PPP1R15B•PP1 complex, preventing it from dephosphorylating eIF2α efficiently. The ISR stays stuck on. Protein synthesis is chronically suppressed. The mice have severe deficits in long-term memory formation.

This was the causal proof: persistent ISR activation, in the absence of any obvious cellular damage, is sufficient to impair memory. The brake on protein synthesis is itself the disease.

Then they turned to DP71L.

Through detailed molecular and structural analysis, they found that DP71L binds PP1 more tightly and more efficiently than the human protein it mimics. It is not merely a viral copy of PPP1R15B — it is a stripped-down, optimized version. Where the human protein is part of a complex regulatory system with many modulators and feedback loops, the viral version is essentially a molecular crowbar that holds the ISR in the off position with unusual force.

The researchers tested DP71L in three different mouse models of cognitive impairment.

In mice modeling Down syndrome — who carry an extra copy of chromosome 21, overexpressing many genes including those that activate the ISR — DP71L reversed the memory deficits. In mice modeling Alzheimer’s disease — where chronic protein misfolding constantly activates cellular stress responses — DP71L reversed the cognitive impairments. In aged mice, where the ISR chronically activates as part of the normal biology of aging, DP71L restored long-term memory to levels seen in young animals.

And in healthy, young mice with no cognitive impairments at all, DP71L enhanced long-term synaptic plasticity and memory beyond baseline.

What This Means — And What It Does Not

Before anything else: this is mouse research. The gulf between a mouse model of Alzheimer’s disease and the human condition is enormous, and many molecules that look miraculous in rodents have failed in clinical trials. Intellectual honesty demands acknowledging that upfront.

But the reason this study is being discussed seriously is the mechanistic clarity of the finding. The researchers are not proposing a vague “neuroprotection” effect. They have a precise molecular explanation: DP71L inhibits eIF2α phosphorylation, restoring protein synthesis, which allows memory consolidation to proceed. The connection between the molecular mechanism and the behavioral outcome is direct and tested from multiple angles.

The fact that it works across three very different models of cognitive impairment — Down syndrome, Alzheimer’s, aging — is particularly striking. These conditions have different primary causes: trisomy 21, amyloid pathology, and the accumulated stresses of time. But they all converge on the same pathway. They all keep the ISR chronically active. And blocking the ISR seems to restore function in all three cases.

This is why researchers have started calling the ISR a potential “common entry point” for cognitive aging and neurodegeneration.

There are real questions to answer before anything like DP71L reaches a clinic. How would you deliver it to the brain? (The team used viral gene delivery in these experiments, which has its own safety and specificity challenges.) What are the consequences of chronically suppressing the ISR — a system that exists for good reasons? What happens in long-term treatment? Are there tumorigenic risks from suppressing a key cellular quality-control pathway?

These are serious questions. But they are the right kind of questions to be asking: specific, answerable, pointing toward a clear experimental path.

Evolution as Drug Discovery

There is something wonderful about the conceptual architecture of this story. The integrated stress response is ancient — it appears in everything from yeast to humans. Viruses have been competing against it for longer than multicellular animals have existed. In that arms race, evolution has sometimes produced molecular tools of extraordinary precision, honed by billions of years of pressure to defeat a specific biological mechanism.

Scientists are increasingly recognizing that the viral proteome — the enormous library of proteins evolved by viruses to manipulate host cells — is a potential treasure chest of biological tools. CRISPR-Cas9 itself came from a bacterial immune system. Viral vectors are the workhorses of gene therapy. And now, a protein from one of the world’s most feared pig pathogens may point the way toward treating Alzheimer’s disease.

DP71L was not designed by anyone. It evolved because African Swine Fever Virus needed a way to survive inside a cell that was doing everything in its power to resist infection. But in the process, it created something that humanity might one day use to preserve memory, to slow the cognitive losses of aging, and to address conditions that affect hundreds of millions of people worldwide.

The pig virus did not know it was solving one of neuroscience’s biggest puzzles. But it did.

---

Footnotes

1. Reineke LC, Zhu PJ, Dalwadi U, et al. “Harnessing viral strategies to reverse cognitive dysfunction through the integrated stress response.” Science 388, (2026). DOI: 10.1126/science.aea8782 ↩ ↩²

2. Detailed review of the ISR-memory connection in: Costa-Mattioli M & Walter P. “The integrated stress response: From mechanism to disease.” Science 368, eaat5314 (2020). DOI: 10.1126/science.aat5314 ↩

3. Sidrauski C, McGeachy AM, Ingolia NT, et al. “The small molecule ISRIB reverses the effects of eIF2α phosphorylation on translation and stress granule assembly.” eLife 4, e05033 (2015). DOI: 10.7554/elife.05033 ↩