If this were a movie, neurons would get top billing as jittery stars firing off electrical one-liners, olanzapine would enter as the mysterious fixer, and astrocytes would be the stagehands everyone ignored until the chandelier started swinging. In Kleefstra syndrome, the plot begins with one underpowered copy of EHMT1, a gene that helps tune which genes stay quiet and which get to bang cymbals at breakfast. When that timing system slips, development can wobble. For some people, skills once gained can regress during puberty or young adulthood, like a clock losing teeth from a gear it already spent years learning to turn.

The Gene With the Tiny Wrench

Kleefstra syndrome is usually caused by a missing or disabled copy of EHMT1. MedlinePlus and GeneReviews describe a multisystem condition involving developmental delay, limited speech, weak muscle tone, seizures, and sometimes adolescent apathy or catatonia-like unresponsiveness. So this is not just "a gene thing," as if genes were decorative drawer labels. EHMT1 helps regulate chromatin, the spool-and-thread packaging system that decides how DNA gets read. Pull the wrong lever and the cellular clockwork starts ticking out of rhythm.

A 2024 study in The American Journal of Human Genetics examined 209 people with rare EHMT1 variants and confirmed Kleefstra syndrome molecularly in 191, broadening the known range of symptoms and variant effects (doi:10.1016/j.ajhg.2024.06.008, PMC11339614). Translation: the condition is more varied than one tidy diagram can admit, because biology enjoys making diagrams look smug.

Plot Twist: The Support Cells Have Lines

The new JCI study asked whether olanzapine, an antipsychotic already used in psychiatry, might help with Kleefstra-associated regression. In an international cohort of 54 people with Kleefstra syndrome, 16 had been treated with olanzapine. Ten showed beneficial improvement in adaptive functioning, and four showed temporary improvement (doi:10.1172/JCI195803). That is intriguing, but it is not a victory parade with confetti cannons. It is observational evidence, and rare-disease cohorts are small enough that statistics have to wear bifocals.

The clever part is what came next. The researchers used human induced pluripotent stem cell-derived neural systems and cortical slices from a Kleefstra mouse model. They found that EHMT1+/- neuronal networks became hyperactive when grown with EHMT1+/- astrocytes. The neurons were not merely being dramatic, although neurons do have theater-kid energy. Their astrocyte neighbors seemed to help wind the network too tightly.

S100B: Small Name, Big Racket

Astrocytes are often described as support cells, which makes them sound like they bring snacks and label cables. In reality, they tune synapses, regulate ions and neurotransmitters, feed neurons, shape inflammation, and keep the brain from becoming a badly assembled cuckoo clock. A 2023 review in Signal Transduction and Targeted Therapy frames astrocytes as central players in neurological and neuropsychiatric disease, sometimes protective, sometimes harmful, and highly context-dependent (doi:10.1038/s41392-023-01628-9). A 2022 Nature Neuroscience study also showed that abnormal astrocyte-secreted proteins can alter neuronal development across several neurodevelopmental disorder models (doi:10.1038/s41593-022-01150-1, PMC10395413).

In this JCI paper, the suspicious astrocyte signal was S100B, a calcium-binding protein often associated with glial activity and neuroinflammation. EHMT1+/- astrocytes had elevated S100B. Olanzapine reduced S100B levels. Even better, blocking S100B pharmacologically or knocking it down genetically in EHMT1+/- astrocytes was enough to rescue the neuronal hyperactivity phenotype. That is the sort of mechanistic clue scientists love because it points to a gear, not just a rattling sound somewhere inside the machine.

Why This Matters, Carefully

For families facing regression, "wait and see" can feel like watching a clock run backward while everyone argues about the battery. Patient advocacy groups describe regression as loss of cognitive, language, motor, adaptive, or self-help skills, often after puberty. One IDefine story captures the uncertainty families can face while clinicians rule out seizures, autoimmune disease, psychiatric symptoms, and everything else in the diagnostic junk drawer.

If these findings hold up, they suggest two useful ideas. First, Kleefstra regression may involve network overactivity shaped by astrocytes, not only neurons. Second, a medication already sitting on pharmacy shelves might point toward a treatable pathway. That does not mean olanzapine is suddenly a universal answer. It can cause major metabolic side effects, including weight gain, and any use in Kleefstra syndrome belongs in specialist care, not in a "saw it online, adjusted the dial myself" situation.

Meanwhile, the field is moving upstream. In June 2026, IDefine and UT Southwestern announced a preclinical collaboration to explore EHMT1 gene replacement, with safety and proper gene dosage as major concerns. That is a longer-range project, more Swiss watch repair than duct tape. The JCI work sits closer to the present: can overactive circuitry be calmed by targeting the astrocyte-S100B gear?

In a rare condition where timing can mean the difference between holding onto a skill and watching it slip, sharper tools matter.

References

1. Vermeulen-Kalk K, Wang S, Kummeling J, et al. Astrocytes contribute to olanzapine-mediated reversal of Kleefstra Syndrome-associated neurodevelopmental regression. J Clin Invest. 2026. doi:10.1172/JCI195803
2. Rots D, Bouman A, Yamada A, et al. Comprehensive EHMT1 variants analysis broadens genotype-phenotype associations and molecular mechanisms in Kleefstra syndrome. Am J Hum Genet. 2024;111(8):1605-1625. doi:10.1016/j.ajhg.2024.06.008, PMC11339614
3. Caldwell ALM, Sancho L, Deng J, et al. Aberrant astrocyte protein secretion contributes to altered neuronal development in multiple models of neurodevelopmental disorders. Nat Neurosci. 2022;25(9):1163-1178. doi:10.1038/s41593-022-01150-1, PMC10395413
4. Verkhratsky A, Butt A, Li B, et al. Astrocytes in human central nervous system diseases: a frontier for new therapies. Signal Transduct Target Ther. 2023;8:396. doi:10.1038/s41392-023-01628-9
5. Kleefstra T, de Leeuw N. Kleefstra Syndrome. GeneReviews. Updated January 26, 2023. NCBI Bookshelf

Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.