In fatal genetic diseases like Huntington’s and spinocerebellar ataxia, proteins develop long stretches of repeating letters that are prone to sticking together like Velcro.

As the proteins clump up in the brain, they damage and kill neurons, triggering severe cognitive decline. Clinicians don’t have any treatments targeting these disease-causing clumps, and Huntington’s and related diseases have no cure.

But there’s a mystery: Some proteins naturally have the same repetitive regions — called polyglutamine or polyQ expansions — that are just as long, or even longer, and they don’t clump at all.

A study from Stanford Medicine, published in Cell in April, dives into that mystery. The research team focused on FOXP2, a gene famously tied to human speech and language.

To their surprise, the protein contains one of the longest known polyQ stretches in the human body, but it usually doesn’t clump or cause disease.

“We got really excited about figuring out why this was,” says Joanna Wysocka, PhD, the Lorry Lokey Professor and a professor of chemical and systems biology and of developmental biology. “Could this protein teach us anything about how to stop the proteins in Huntington’s from clumping together?”

It’s the kind of basic science question that seemed obscure at first but turned out to have major implications in treating disease.

Shady Saad, seated, Daniel Jarosz, and Joanna Wysocka examine an image of FOXP2 protein in the Wysocka Lab. (Photo by Jim Gensheimer)

What’s so special about FOXP2?

FOXP2 is often nicknamed the “speech gene.” Mutations in FOXP2 have been linked to verbal dyspraxia, a condition in which people can understand language but struggle to coordinate the movement of lip, tongue and jaw needed for speech. Scientists have hypothesized that changes in FOXP2 may explain how humans first evolved the ability to speak.

FOXP2 is nearly identical in most mammals, but there are two tiny characteristics found only in humans. When mice have the human version of the gene, they make more complex vocalizations and can more easily form connections in the brain.

The clumping connection

Huntington’s is one of nine known diseases caused by long polyQ expansions, in which three letters of DNA repeat over and over. When the Huntington’s gene has more than 36 polyQ repeats in a row, it usually causes disease. Symptoms typically appear in mid-adulthood and include involuntary movements, memory problems and mood changes. The disease is progressive and always fatal.

Because of its role in speech, we knew that FOXP2 is active in neurons. Why does the brain tolerate this long polyQ stretch but not the shorter polyQ stretch seen in Huntington's?"

— Joanna Wysocka

FOXP2, surprisingly, has a natural polyQ stretch with more than 40 repeats in a row, followed by 10 more repeats nearby — but it doesn’t clump. The Stanford Medicine team wanted to understand why.

“Because of its role in speech, we knew that FOXP2 is active in neurons,” Wysocka explained. “Why does the brain tolerate this long polyQ stretch but not the shorter polyQ stretch seen in Huntington’s?”

This question had fascinated Shady Saad, PhD, a postdoctoral fellow who is mentored by Wysocka and Daniel Jarosz, PhD, an expert on protein aggregation and an associate professor of chemical and systems biology and of developmental biology.

“This protein has such an extreme sequence that from the outset it seemed likely multiple mechanisms would be needed to prevent its aggregation,” Jarosz said.

The mysteries of FOXP2

How might the speech gene's anti-clumping properties help combat neurodegenergative diseases such as Huntington's?

Photos by Jim Gensheimer
Lab process detailed by Shady Saad

Shady Saad looks at a structure of an aggregation-prone human protein in the Wysocka Lab, Lokey Stem Cell Research Building, at Stanford University, on Wednesday, August 20, 2025, in Stanford, Calif. (Photo by Jim Gensheimer) This image represents one of the 5 steps in the lab’s approach to their science. 1) Used computer search methods to look for endogenous human proteins with aggregation prone sequences and then to analyze the structure and evolution of FOXP2.

1. DISCOVERY: Researchers first use computer search methods to look for endogenous human proteins with aggregation prone sequences and then to analyze the structure and evolution of FOXP2.

Shady Saad prepares protein samples for analysis in the Wysocka Lab, Lokey Stem Cell Research Building, at Stanford University, on Wednesday, August 20, 2025, in Stanford, Calif. (Photo by Jim Gensheimer) This image represents one of the 5 steps in the lab’s approach to their science. 2) Preparing different protein fusions to test how DNA binding and phosphorylation affect clumping of FOXP2 and Huntingtin.

2. PREPARATION: Different protein fusions are then prepared to test how DNA binding and phosphorylation affect clumping of FOXP2 and Huntington's protein.

Shady Saad introduces different aggregation-prone proteins to human cells in the Wysocka Lab, Lokey Stem Cell Research Building, at Stanford University, on Wednesday, August 20, 2025, in Stanford, Calif. (Photo by Jim Gensheimer) This image represents one of the 5 steps in the lab’s approach to their science. 3) Cell cultures — added fused FOXP2 and Huntington's protein to cells to test effect.

3. FUSING: Fusions of FOXP2 and Huntington's proteins to solubilizing tags were then introduced to cells to test the effect.

Shady Saad looks through a microscope that captures digital images in the Wysocka Lab, Lokey Stem Cell Research Building, at Stanford University, on Wednesday, August 20, 2025, in Stanford, Calif. (Photo by Jim Gensheimer) This image represents one of the 5 steps in the lab’s approach to their science. 4) Microscopy — used many different labeling and microscopy methods to watch how the proteins clumped in living cells. The proteins are then visualized in living cells using microscopy

4. VISUALIZATION: Different labeling and microscopy methods are used to watch how the proteins clump in living cells. The proteins are then visualized in living cells under a microscope.

Shady Saad, at left, analyses FOXP2 protein from different species with Joanna Wysocka, and Raquel Fueyo in the Wysocka Lab, Lokey Stem Cell Research Building, at Stanford University, on Wednesday, August 20, 2025, in Stanford, Calif. (Photo by Jim Gensheimer) This image represents one of the 5 steps in the lab’s approach to their science. 5) Comparing FOXP2 across species. They used some gels in this section.

5. COMPARISON: The researchers then compare FOXP2 across species to make a determination.

FOXP2’s protective factor

Through a series of experiments in the lab, Saad uncovered two main reasons FOXP2 usually doesn’t clump:

It binds to DNA. FOXP2 is a type of protein that binds to DNA to regulate genes. Indeed, in most cells, FOXP2 is bound to DNA, which seems to help spread it out — preventing it from piling up and sticking to itself. When the protein residues responsible for FOXP2 binding to DNA are perturbed — as seen in verbal dyspraxia — the protein starts to clump. “We think that DNA, which is a stiff polymer, and potentially other factors attached to it help disperse FOXP2, such that individual proteins have less of a chance of coalescing and forming aggregates,” Wysocka explained.
It gets a charged chemical coat during cell division. When cells start dividing, most transcription factors detach from DNA. So why doesn’t FOXP2 start sticking together at that point? During this phase, it turns out, the protein gains another anti-clumping protection: Chemical groups called phosphates are added to its structure. This negatively charged chemical coating prevents FOXP2 protein molecules from sticking together. And when the ability of FOXP2 to undergo phosphorylation is blocked, the protein again starts clumping, just like the proteins involved in disease.

Could this help treat disease?

Together, DNA binding and phosphorylation act like natural anti-clumping systems for FOXP2. The researchers wondered whether they could use the same strategies to tackle pathogenic proteins associated with polyQ expansion diseases. “Since FOXP2 and Huntington’s share the same aggregation signature, we asked if our anti-clumping mechanisms could also solubilize the pathological protein,” Saad explained.

To test that idea, the team fused the clump-prone Huntington’s protein with FOXP2’s protective features. Amazingly, those additions were enough to reduce — or in some cases fully dissolve — the toxic clumps. Their findings included:

Adding a DNA-binding tag to Huntington’s proteins helped break them apart inside cells.
Mimicking the negatively charged phosphate molecules found on FOXP2 also made a difference, preventing sticky aggregates from forming.
Even the tough insoluble aggregates called amyloids that mimic those already formed in Huntington’s patients became more soluble with these add-ons.

“I was very surprised that it worked like a charm,” Wysocka said. “These are notoriously hard-to-break aggregates. The fact that they started to dissolve was really exciting.”

“It was astonishing to find that our cells already have the ability to dissolve these aggregates,” Jarosz said. “It was hidden in plain sight.”

“The results were black and white, the amyloids of polyQ were gone, it was a happy moment,” Saad said.

BEHIND THE SCIENCE

Ribosomes, the protein factories of cells

Addiction

Lessons in evolution

The story came full circle when the team compared FOXP2 proteins from humans, chimpanzees and mice. They found that the human version was significantly more soluble — less prone to clumping — than versions from other species.

It turned out that the same changes to FOXP2 that boost vocalization and brain flexibility in mice also make the protein less prone to clumping. In other words, the evolutionary tweaks that helped humans develop speech may made FOXP2 safer at higher levels in the brain.

“It’s tempting to speculate that this increase in solubility allowed our brains to ramp up FOXP2 levels without causing harm,” Wysocka said. “That might have helped pave the way for human speech to evolve.”

What’s next with FOXP2?

The team is now testing whether their lessons from FOXP2 could help treat Huntington’s and other polyQ diseases. Ultimately, they would like to design drugs that mimic the anti-clumping effects of DNA binding and phosphorylation.

“We have developed a completely new approach to target these diseases by addressing the root cause,” Saad said. “I am very excited for the potential outcome.”

“We started out studying a basic science question about evolution,” Wysocka said. “Now, we’re thinking about how to treat a disease with no current cure. That’s the power of basic research — it can take you to places you never expected.”

* * *

ABOUT THIS SERIES

It's up to basic science — and the researchers who devote their labs and lives to it — to make the discoveries that lead to medical interventions. Behind the Science is devoted to that science and those researchers at Stanford Medicine.