Disorders of GABRA3, the only epilepsy-associated GABAA receptor subunit gene on the X chromosome, have eluded clinical clarity due to ambiguous inheritance patterns and variable phenotypes. The long-standing assumption that all pathogenic variants cause loss of function further obscured genotype–phenotype relationships and hindered progress. Here, we curated a cohort of individuals with a GABRA3 variant, integrating deep phenotyping, genotyping, family history, and electrophysiology with a targeted mouse model. Among 43 individuals with 19 GABRA3 variants, functional analyses revealed gain- and loss-of-function effects, each linked to distinct clinical profiles. Gain-of-function variants were associated with severe, treatment-resistant epilepsy and profound intellectual disability, disproportionately affecting males, who were often nonambulant and had cortical visual impairment. Loss-of-function variants produced milder phenotypes, with epilepsy rarely observed; affected males showed behavioral issues and language delay, while females were unaffected carriers. Our gain-of-function (Gabra3Q242L/+) mouse model mirrored these sex-specific differences, showing increased seizure susceptibility, early death, and marked cortical hyperexcitability. These insights resolve longstanding uncertainties surrounding GABRA3 and redefine how X-linked disorders are interpreted. They demonstrate that it is the functional impact of a variant, not its mere presence, that determines whether a condition manifests dominantly or recessively. This distinction carries important implications for genetic counseling, precision medicine, and the broader interpretation of X-linked neurodevelopmental disorders.
Katrine M. Johannesen, Khaing Phyu Aung, Vivian W.Y. Liao, Nathan Absalom, Han C. Chua, Xue N. Gan, Miaomiao Mao, Chaseley E. McKenzie, Hian M. Lee, Sebastian Ortiz, Rebecca C. Spillmann, Vandana Shashi, Rodney A. Radtke, Ghayda M. Mirzaa, P. Anne Weisner, Josue Flores Daboub, Caroline Hagedorn, Pinar Bayrak-Toydemir, Desiree DeMille, Jian Zhao, Nandita Bajaj, Yline Capri, Boris Keren, Miriam Schmidts, Ingrid M.B.H. van de Laar, Marjon A. van Slegtenhorst, Rafal Ploski, Marta Bogotko, Danielle K. Bourque, Ebba Alkhunaizi, Lauren Chad, Nada Quercia, Houda Elloumi, Ingrid M. Wentzensen, Michael C. Kruer, Pritha Bisarad, Carolina I. Galaz-Montoya, Violeta Rusu, Dominique Braun, Katie Angione, Jessica C. Win, Camilo Espinosa-Jovel, Pia Zacher, Konrad Platzer, Samuel F. Berkovic, Ingrid E. Scheffer, Mary Chebib, Guido Rubboli, Rikke S. Møller, Christopher A. Reid, Philip K. Ahring
Usage data is cumulative from November 2025 through September 2026.
| Usage | JCI | PMC |
|---|---|---|
| Text version | 3,678 | 602 |
| 1,255 | 103 | |
| Figure | 1,032 | 0 |
| Table | 188 | 0 |
| Supplemental data | 1,115 | 48 |
| Citation downloads | 336 | 0 |
| Totals | 7,604 | 753 |
| Total Views | 8,357 | |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.