At Kanazawa University, researchers have uncovered a hidden layer of brain architecture that could reshape how we understand autism spectrum disorder: microscopic nets surrounding neurons in the cerebellum that, when disrupted, impair the very circuits that allow us to connect socially with others.
The discovery matters because autism spectrum disorder affects social interaction and communication in profound ways, yet scientists have struggled to pinpoint exactly how structural changes in the brain drive these challenges. For decades, the cerebellum was dismissed as merely the brain's motor control center—the part that keeps us coordinated and balanced. But in recent years, researchers have recognized it as a crucial regulator of higher-order functions like emotion, cognition, and social behavior. The molecular mechanism linking cerebellar changes to autism's social deficits, however, remained frustratingly unclear.
The Kanazawa team investigated two distinct ASD mouse models that represent different autism risk pathways: one exposed to prenatal valproic acid, an environmental risk factor, and another carrying a mutation in the Chd8 gene, a genetic risk factor. Despite these different origins, both models revealed the same striking finding: neurons in the deep cerebellar nuclei—the cerebellum's major output region—showed a marked reduction in perineuronal nets (PNNs), specialized extracellular matrix structures that wrap around neurons like protective lattices.
These PNNs are not mere scaffolding. They stabilize neuronal excitability, regulate synaptic signaling, and support the maturation of neural circuits. To understand what happens when they vanish, the researchers used an enzymatic approach to selectively degrade PNNs in the cerebellar nuclei of normal mice. The results were clear and sobering: mice with disrupted PNNs displayed impaired social behavior, showing reduced social interaction and decreased interest in unfamiliar mice. The structural change directly caused the behavioral change.
Further investigation revealed the circuit-wide consequence of this loss. Normally, social stimuli activate neurons in the cerebellar nuclei, and this activity ripples outward to distant brain regions including the midbrain and thalamus. But in mice with disrupted PNNs, these cerebellar neurons barely activated at all, and neuronal activity across the entire cerebellum-connected circuit plummeted. It was as if removing these tiny nets silenced a conversation happening across distant parts of the brain.
The researchers identified a molecular culprit: a transcription factor called ARNT2. When PNNs disappeared, ARNT2 expression surged in the affected neurons, pushing them into a less responsive state. But here's where the story offers a glimmer of possibility: when the researchers suppressed ARNT2, both neuronal activity and social behavior bounced back. One molecular switch restored function.
The findings, published in Translational Psychiatry, suggest that autism-linked social difficulties may arise not from isolated brain damage but from disrupted communication across distributed neural circuits. The cerebellum, once considered a minor player in social cognition, emerges as essential. Understanding how PNNs maintain these circuits opens new pathways for thinking about intervention—not fixing a single broken part, but restoring the delicate structures that let our brains work as an integrated whole.