The treatment restores neuronal structure in a way that allows normal circuit function to recover
For decades, magnetic brain stimulation has lifted depression in patients who found no relief in medication, yet the cellular reasons for its success remained hidden from science. Researchers at UCLA Health have now illuminated that mystery, demonstrating in a Cell study how a compressed, five-day stimulation protocol physically rebuilds the synaptic architecture that chronic stress dismantles — and does so with a selectivity that surprised even the scientists conducting the work. The discovery places psychiatric medicine at a threshold where treatment may one day be calibrated not to a diagnosis, but to the precise circuitry of an individual mind.
- Millions of depression patients unresponsive to medication have relied on magnetic brain stimulation without anyone fully understanding why it works — a gap that has limited the therapy's refinement and reach.
- UCLA researchers partnered with the NIH to stimulate the brains of stress-exposed mice in real time, watching chronic stress strip away the tiny dendritic spines that allow neurons to speak to one another.
- A single day of accelerated theta burst stimulation restored those lost connections within 24 hours — but only in one precise class of neurons called intratelencephalic cells, leaving neighboring neurons largely untouched.
- When scientists blocked those specific neurons during treatment, the antidepressant effects disappeared entirely, confirming that the therapy works by structurally repairing a targeted circuit, not by broadly exciting the brain.
- The restored connections held for at least a week after just one day of stimulation, suggesting genuine neural repair rather than temporary relief — and pointing toward personalized protocols for depression, PTSD, OCD, and chronic pain.
For years, clinicians have watched magnetic coils placed against the scalp lift depression in patients who found no relief in pills. They knew it worked. What they could not see was why — the mechanism remained locked inside a cellular black box.
A team at UCLA Health has now opened that box. Publishing in Cell, researchers from the Neuromodulation Division demonstrated for the first time how a fast-acting form of magnetic stimulation physically rebuilds the neural circuits that chronic stress tears apart. The work was led by Dr. Scott Wilke and Dr. Laura DeNardo, working alongside scientists at the National Institutes of Health to develop a method for stimulating awake mice in a way that mirrors clinical treatment.
The therapy, called accelerated intermittent theta burst stimulation or aiTBS, compresses what is normally a six-week treatment course into just five days. Patients have reported rapid relief, but the cellular story behind that relief was unknown — until now. When the team exposed mice to chronic stress and then applied the magnetic pulses, they found that stress had caused neurons in the prefrontal cortex to lose dendritic spines, the tiny protrusions essential for communication between brain cells. After just one day of aiTBS, those lost connections re-emerged — but only within a specific class of neurons called intratelencephalic, or IT, neurons. Neighboring cells were largely unaffected.
"We initially thought TMS might broadly affect the prefrontal cortex, but instead the effects were surprisingly precise," said first author Michael Gongwer, an MD-PhD student at UCLA Health. The precision proved decisive: when researchers selectively blocked IT neuron activity during stimulation, the antidepressant effects vanished entirely, confirming these neurons are not peripheral players but the essential mechanism of repair.
The structural improvements appeared within 24 hours and persisted for at least a week after a single day of treatment. "This isn't just a temporary shift in activity," Wilke noted. "The treatment restores neuronal structure in a way that allows normal circuit function and behavior to recover." The findings carry implications well beyond depression — TMS is already being explored for chronic pain, OCD, PTSD, and tinnitus, all rooted in specific circuit dysfunction. Understanding how stimulation selectively repairs those circuits opens the door to therapies tailored not to a diagnosis, but to the particular wiring that has gone wrong in each patient.
For years, doctors have known that a magnetic coil placed against the scalp can lift depression in people who don't respond to pills. They've watched it work. But they couldn't see why. The mechanism lived in a black box—effective, but mysterious at the level of actual brain cells and the connections between them.
Researchers at UCLA Health have now opened that box. In a study published in Cell, a team from the Neuromodulation Division demonstrated for the first time how a fast-acting form of magnetic stimulation physically rebuilds the neural circuits that chronic stress tears apart. The work was led by Dr. Scott Wilke, an assistant professor of psychiatry and psychiatrist at UCLA's TMS Clinical and Research Service, alongside Dr. Laura DeNardo, an associate professor of physiology at the David Geffen School of Medicine.
The therapy in question is called accelerated intermittent theta burst stimulation, or aiTBS. Unlike standard magnetic stimulation protocols that require daily treatments for six weeks or longer, aiTBS compresses the entire course into five days. Clinicians have begun using it because patients report rapid relief from depressive symptoms. But until now, nobody understood what was actually happening inside the brain to produce those results.
To find out, the UCLA team partnered with scientists at the National Institutes of Health to develop a novel method for stimulating the brains of awake mice in a way that mirrors clinical treatment. They exposed the animals to chronic stress to simulate depression, then delivered the magnetic pulses while monitoring brain activity in real time. What they discovered was striking in its precision. Chronic stress had caused neurons in the prefrontal cortex to lose dendritic spines—tiny protrusions that are essential for communication between brain cells. This structural damage appeared across multiple neuron types. But when the researchers applied just one day of aiTBS, something unexpected happened. The lost connections re-emerged, and activity surged—but only in one specific class of neurons called intratelencephalic, or IT, neurons. The neighboring cells largely remained unchanged. "We initially thought TMS might broadly affect the prefrontal cortex, but instead the effects were surprisingly precise," said Michael Gongwer, the study's first author and an MD-PhD student at UCLA Health.
The precision proved crucial. When the researchers selectively blocked IT neuron activity during stimulation, the antidepressant effects vanished entirely. This demonstrated that these neurons are not just incidentally involved—they are essential. The stimulation works by restoring the structural scaffolding that stress had dismantled, allowing these specific neurons to re-engage circuits that support adaptive behavior. The improvements in stress-related behaviors appeared within 24 hours of treatment and persisted for at least a week after a single day of stimulation, accompanied by stable structural changes in the IT neurons themselves.
"What's striking is that this isn't just a temporary shift in activity," Wilke said. "The treatment restores neuronal structure in a way that allows normal circuit function and behavior to recover." The durability of the effect—lasting weeks from just five days of treatment—suggests that the brain is not merely being jolted into temporary relief but is being fundamentally repaired. While animal models cannot fully capture the complexity of human depression, this research provides some of the strongest evidence yet for how brain stimulation produces therapeutic effects at the cellular level.
The implications extend beyond depression. TMS is already being tested for chronic pain, obsessive-compulsive disorder, post-traumatic stress disorder, and tinnitus—all conditions rooted in dysfunction within specific brain circuits. Understanding how stimulation can selectively target and repair those circuits opens a path toward more effective treatments. "Every patient is unique," Wilke noted. "By studying these treatments in mice, we can systematically test how different stimulation parameters reshape brain circuits, which may ultimately help us tailor neuromodulation therapies to individual patients." The work suggests a future where magnetic brain stimulation becomes not a one-size-fits-all intervention but a precision tool, calibrated to the particular circuitry that has gone wrong.
Notable Quotes
We initially thought TMS might broadly affect the prefrontal cortex, but instead the effects were surprisingly precise.— Michael Gongwer, first author and MD-PhD student at UCLA Health
By studying these treatments in mice, we can systematically test how different stimulation parameters reshape brain circuits, which may ultimately help us tailor neuromodulation therapies to individual patients.— Dr. Scott Wilke, assistant professor of psychiatry at UCLA Health
The Hearth Conversation Another angle on the story
So the magnetic coil was always working, but nobody knew why?
Exactly. Clinicians could see the results in their patients—depression lifting, sometimes quickly—but the actual mechanism at the cellular level was a complete mystery. It was like having a key that opens the lock but not understanding how the mechanism inside works.
And this study shows that the stimulation is actually rebuilding physical structures in the brain?
Yes. Chronic stress damages these tiny protrusions called dendritic spines that neurons use to talk to each other. The magnetic stimulation doesn't just activate neurons—it restores those physical connections. The spines actually grow back.
But it only affects certain neurons, not all of them?
That's the remarkable part. It targets a very specific class of neurons called IT neurons in the prefrontal cortex. When the researchers blocked just those neurons, the whole therapeutic effect disappeared. So the brain isn't being broadly stimulated—it's being precisely repaired.
How long does the effect last?
They saw improvements within 24 hours and the benefits persisted for at least a week after just one day of treatment. That's remarkable given how brief the intervention is. The structural changes in the neurons were stable, which suggests the brain isn't just being temporarily activated—it's being fundamentally rebuilt.
Does this mean we could eventually personalize these treatments?
That's the hope. If we understand which circuits are disrupted in a particular patient and how different stimulation parameters affect those circuits, we could theoretically tailor the treatment to that person's specific brain dysfunction. Right now it's one protocol for everyone. This research points toward something much more precise.