SAN FRANCISCO— Whether focusing on a task, learning a new skill, or retrieving a cherished memory, the human brain relies on rapid, rhythmic electrical pulses known as gamma oscillations. For decades, scientists knew these rhythms were weakened in Alzheimer’s disease. But how individual brain cells coordinate their activities to generate them—and why that process fails—remained a mystery.
Now, researchers at Gladstone Institutes have unlocked a key piece of the puzzle.
In a study published in Neuron, scientists discovered that specialized rhythm-keeping cells called parvalbumin (PV) interneurons generate these brain rhythms by taking precise turns firing. But in mouse models of Alzheimer's disease, this delicate coordination collapses, causing memory waves to falter.
Critically, by boosting levels of a specific protein that helps these PV cells generate electrical pulses, the researchers were able to restore the normal rhythm and brain activity, offering a promising therapeutic target for restoring cognitive function.
“Gamma oscillations are essential for clear thinking and memory, but we've lacked a complete picture of how the brain creates them,” says Jorge J. Palop, PhD, a Gladstone investigator and senior author of the study. “What makes this discovery so exciting is that we didn't just figure out how healthy cells pass the baton; we discovered a way to reset their timing when Alzheimer's related processes throw them off beat.”
Catching Cells in Action
Gamma oscillations are fast, repetitive brain rhythms—cycling 30 to 90 times a second—that happen in the brain’s background during tasks requiring attention and memory. Previously, scientists relied on methods like optogenetics to activate all PV interneurons simultaneously to demonstrate that PV cells generate gamma rhythms. However, broad stimulation couldn't reveal how individual cells behave naturally in real time.
“Turning on all the cells at once shows you what they can do, but not how they naturally work together or how they’re recruited to generate spontaneous gamma oscillations,” says first author Keran Ma, PhD, who led the work as a postdoctoral scholar at Gladstone and is now an assistant professor at UTHealth Houston. “Until now, no one has actually been able to image these cells and look at gamma oscillations at the same time in an awake animal during natural behavior.”
To overcome this obstacle, Gladstone scientists developed an innovative technique combining real-time electrical recording with imaging of live cells. By monitoring mice as they ran on a treadmill—a state that naturally prompts attention-related brain rhythms—the researchers engineered PV cells to glow fluorescently every time they fired an electrical pulse.
The images revealed an intricate relay race: every time a burst of gamma oscillations began, PV cells became active not all at once, but in successive waves. The first cells to respond showed high activity, followed by subsequent groups with progressively weaker firing, while roughly a third of the cells remained quiet—demonstrating a highly organized pattern of turn-taking essential for healthy brain function.
“The relationship between the PV cells and the gamma oscillations was much more dynamic and complex than we expected going into this,” Palop says. “Rather than all the cells contributing to the rhythm at once, each cell contributes for just a brief moment, and those small contributions add up to create the larger brain rhythm. It’s less like a military lockstep march and more like a flock of birds in the sky—each responding to the others, shifting turns and trajectories, yet together creating a remarkably coordinated pattern.”
Off-Beat in Alzheimer’s
The team then studied mice engineered to develop Alzheimer’s-like disease. In these mice, the rhythm’s timing broke down; PV cells that fired first were sluggish, those that joined late were overactive, and fewer cells stayed quiet. This resulted in too much synchrony—more cells fired at the same time and level, and gamma rhythms became weaker.
“The important thing here is that more activity is not always going to produce better oscillations or better cognition,” Ma says. “The dynamics of the cells are most important.”
The findings could help explain why more than a quarter of people with Alzheimer’s also show epilepsy-like activity in the brain. In the mice involved in the study, epileptic spikes appeared after sudden drops in PV interneuron activity.
A Path Toward Treatments
Palop’s lab previously found that levels of Nav1.1—a sodium channel found on the surface of PV cells that helps them fire—are reduced in both Alzheimer’s mouse models and people with the disease. In the new study, the team engineered the Alzheimer’s mice to produce more Nav1.1 and then watched their PV cells and brain activity.
The extra channels restored normal waves of PV interneuron activity and strengthened gamma oscillations. They also reduced the epileptic activity, reduced amyloid plaques (a pathological hallmark of Alzheimer’s), and improved the memory, cognition, and lifespans of the mice. More work is needed to understand the link between PV cell failure and Alzheimer’s in humans, and how Nav1.1 levels or activities could be boosted.
“We still don’t know why interneuron dysfunction happens early in Alzheimer’s disease," Ma says. “But this is another study underscoring that a lot of the pathology of the disease could arise downstream of disrupted interneurons and the abnormal brain rhythms this disruption produces.”
To pinpoint these rhythm-keeping cells, the team relied on innovative genetic tools developed at the Allen Institute. Study co-authors Boaz P. Levi, PhD, and Jonathan T. Ting, PhD, helped design these delivery vehicles to zero in on specific brain cell types across different species—a crucial step toward creating precise therapies for humans.
“One of the most exciting aspects of this work is our ability to target a single cell type out of the hundreds in the brain,” Levi says. “Because these tools work in both mice and humans, they help us bridge the gap between understanding how individual cells fail in disease and developing targeted treatments to fix them.”
“This collaboration illustrates where this field is headed,” Ting adds. “By combining precise delivery tools with a deep understanding of Alzheimer’s, we can start designing treatments that act only on the specific cells responsible for restoring healthy brain function.”
The researchers also still have basic questions about gamma oscillations, such as what dictates which cells are active at which time.
“It could be that different thoughts or brain processes are regulated by different assemblies of PV cells,” Palop says. “A lot of interesting questions remain to be answered in the future as we continue our research.”
In addition to addressing these unanswered questions, Palop says he and his team are using their latest findings to move toward the development of potential therapies. They’re currently working to develop small-molecule Nav1.1 enhancers designed to boost interneuron activity in Alzheimer’s disease and other disorders.
About the Study
The paper, “Parvalbumin cell state dynamics regulate gamma oscillations and are
disrupted in Alzheimer’s disease,” was published in the journal Neuron on October 15, 2026. In addition to Ma and Palop, authors are Kyerl Park, Patrick Honma, Yutong Zhang, Erica Brady, Yuechen Qiu, and Katerina Akassoglou, who conducted this research at Gladstone and is now at Rensselaer Polytechnic Institute and the Icahn School of Medicine at Mount Sinai; and Ximena Opitz-Araya, Jonathan T. Ting, and Boaz P. Levi of the Allen Institute.
The work was supported by the Ray and Dagmar Dolby Family Fund, the Cure Alzheimer's Fund, the Simon Family Trust, the Alzheimer’s Association (24AARG-1247332), the Texas Alzheimer’s Research and Care Consortium (1271368), and the National Institutes of Health (AG094667, AG092683, AG082147, AG062629, AG073082, UG3/UH3 MH120095, R35 NS097976, R35 NS143067, RF1AG064926, K99/R00AG062776).
Neuron
Parvalbumin cell state dynamics regulate gamma oscillations and are disrupted in Alzheimer’s disease
15-Oct-2026