The investigation emerged from a long-standing question about why males appear more vulnerable to autism. Corresponding author Dr. Huda Zoghbi, Distinguished Service Professor at Baylor, director of the Duncan NRI and a Howard Hughes Medical Institute investigator, describes autism as a common, heritable neurodevelopmental trait that affects how a person communicates, interacts with others and experiences the world, characterized by altered social interactions and restricted, repetitive behaviors or interests. The pronounced sex skew in diagnosis has long hinted that sex-specific factors may raise male susceptibility, but identifying those factors has proven difficult. The new study offers a concrete mechanistic hypothesis rooted in the peculiar genetics of the sex chromosomes.

The logic behind the hypothesis rests on a fundamental difference between male and female genomes. Females carry two X chromosomes, while males carry only one. First author Dr. Rebecca Meyer-Schuman, a postdoctoral associate in the Zoghbi lab, explained the team’s reasoning: mild mutations in the regulatory regions of an X chromosome gene might be tolerated in females, whose second chromosome can buffer the effect, but hit males with full force because they possess only a single copy. Regulatory regions act like dials that control whether a gene is expressed and by how much. If one of those dials is slightly miscalibrated on a male’s only X chromosome, there is no backup copy to compensate.

MECP2 was an obvious starting point for testing this idea. Decades of research, including earlier mouse studies from the Zoghbi lab and other groups, established that the protein produced by MECP2 behaves in what scientists call a Goldilocks fashion: the brain requires a just-right concentration or activity level. Too little of the protein, roughly half of normal levels or less, causes Rett syndrome, a profound neurodevelopmental disorder. Too much, about twice the normal amount, causes MECP2 duplication syndrome, a different condition with its own severe features. The gene sits in an unusually narrow therapeutic and biological window, which makes its regulatory regions especially consequential.

Crucially, the mouse studies also showed that milder shifts in MECP2 dosage fall into a different clinical territory. Reducing expression by about 30 percent or increasing it by 50 percent does not produce the defining characteristics of Rett syndrome or MECP2 duplication syndrome, such as seizures or motor problems. Instead, these moderate changes lead to autism-like behaviors in the animals. That dose-response relationship gave the researchers a precise prediction to test in human genetics: regulatory variants that modestly reduce MECP2 expression should be found in autistic individuals, particularly males, and should not be accompanied by the severe symptoms of the better-known MECP2 disorders.

To interrogate the regulatory regions of MECP2 systematically, the team employed a laboratory technique called a Massively Parallel Reporter Assay, or MPRA. This method allows researchers to test thousands of DNA sequences simultaneously by placing them upstream of a reporter gene and measuring how each sequence drives expression. In this way the team could identify and manipulate the regulatory regions of MECP2 and determine what kinds of sequence changes would alter their function. The assay served as a functional screening platform, converting raw genetic variation into measurable effects on gene activity rather than relying on computational predictions alone.

The researchers then bridged from the bench to human patients. They screened the MECP2 regulatory regions in autistic individuals and took the genetic variants they identified back into the MPRA assays to test their functional consequences. Two regulatory regions emerged where male autistic individuals had inherited a variant from their unaffected mother that altered regulatory activity. In one case, the variant reduced MECP2 expression by approximately 30 percent in human neurons, a magnitude that in mice produces social deficits, hyperactivity and anxiety-like characteristics. Consistent with the mouse model prediction, that individual was diagnosed with autism and ADHD but displayed none of the defining characteristics of Rett syndrome.

The pattern of inheritance is itself informative. Because the variants were carried by unaffected mothers, they fit the expectation that females are buffered against mild MECP2 regulatory changes, likely through the contribution of their second X chromosome, while their sons, receiving a single X chromosome, express the altered dosage. This transmission pattern mirrors the male bias in autism diagnosis and provides a plausible genetic route by which a variant can persist silently in the female population while producing neurodevelopmental effects in males. It also illustrates why X-linked regulatory variation deserves closer attention in studies of autism’s sex skew.

Zoghbi characterized the findings as the result of integrating human and animal studies, suggesting that MECP2 regulatory variants can contribute to male-biased autism while revealing just the tip of the iceberg. In her view, the study provides a framework for uncovering regulatory variants in other X chromosome neurodevelopmental genes that may contribute to autism’s missing heritability, the substantial portion of autism risk that twin and family studies attribute to genetics but that sequencing of protein-coding regions has not yet fully explained. Regulatory regions, which do not encode proteins but control gene activity, are increasingly recognized as a rich hunting ground for such unaccounted risk.

Technically, the study demonstrates the power of pairing high-throughput functional assays with patient genetics. Rather than treating every noncoding variant as a candidate, the MPRA approach assigns functional weight to specific sequences, and the mouse dosage data supply a quantitative bridge between a measured percent change in expression and an expected behavioral outcome. The convergence of all three lines of evidence, functional assay, human variant and animal dose-response, is what elevates these two regulatory regions from statistical associations to mechanistic candidates. Other contributors to the work include Fisher Cherry, Yang Sui, Athanasios Papastathopoulos-Katsaros, Yi Zhong, Yidan Li, Tianyun Wang, Kelsey Hennick, Druha Karunakaran, Hanna Berk-Rauch, Zhandong Liu, Aravinda Chakravarti, Tomasz Nowakowski and Evan E. Eichler. The findings open a path toward systematically scanning the X chromosome for regulatory variants that quietly shape who develops autism, and why boys bear the greater burden.

Subject of Research: X-linked MECP2 regulatory variants and male-biased autism susceptibility

Article Title: New genetic candidates help explain why autism affects males more often than females

Article References: New genetic candidates help explain why autism affects males more often than females. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: autism, MECP2, X chromosome, regulatory variants, male bias, Rett syndrome, MPRA, genetics, neurodevelopment, Baylor College of Medicine, gene expression, missing heritability

Rett syndrome is a neurodevelopmental disorder that affects girls almost exclusively. In the typical course, infants appear healthy at birth and develop normally for the first months of life, only to stagnate and then regress, losing purposeful hand use and spoken language. Characteristic features follow: repetitive hand-to-mouth movements, teeth grinding, episodes of intense crying, constipation, sleep and feeding difficulties, and a slowing of head growth. Many children eventually enter a plateau phase in which some skills, particularly eye-mediated communication, partially recover, but dystonia, scoliosis, breathing irregularities, and epilepsy may emerge. A recent natural history study found that most skill acquisition in Rett syndrome occurs before six years of age, and that when skills are regained, that too generally happens before this age. The genetic culprit in the overwhelming majority of classic cases is MECP2, an X-linked gene encoding an epigenetic regulator that binds methylated DNA and orchestrates the activity of neuronal genes, acting as both a transcriptional repressor and activator.

The Swedish girl at the center of the new report was born in 2009 after an uneventful pregnancy, weighing 2760 grams with a normal head circumference. Yet the first months of life already hinted at trouble. By six months she was hypotonic, with poor trunk stability, delayed motor development, and feeding difficulties. Extensive early investigations, including chromosome analysis, metabolic testing, muscle biopsy, and a mitochondrial workup, all came back inconclusive. By eleven months she lacked direct eye contact, could not sit unsupported, made writhing hand movements, and screamed frequently, requiring tube feeding. At twelve months a gastrostomy was placed, and brain MRI revealed white matter loss most pronounced in the fronto-parietal regions. Sequencing of MECP2 and CDKL5, another gene associated with atypical Rett presentations, was normal, as was a chromosomal microarray that ruled out microdeletions and microduplications.

The clinical picture continued to evolve in a way that would eventually become unmistakable. At fourteen months she developed severe screaming spells, dystonia in both feet, and self-injurious hand biting. By age three she was non-ambulant, without language or eye contact, and was diagnosed with autism. At age seven, following evaluation by an Rett syndrome specialist, she was judged to fulfill the diagnostic criteria for classic Rett syndrome: partial loss of acquired purposeful hand function, loss of acquired spoken language, inability to walk, and stereotypic hand movements. She could sit unsupported, communicate with smiles and eye gaze, and return a rolling ball, but she also displayed respiratory dysfunction with intense hyperventilation and breath-holding, bruxism, and stereotypic hand-mouth movements. Now fifteen, she struggles most with behavioral problems, including cyclical screaming spells and self-injury, attending special school four to five days a week during good phases and only two days during bad ones.

To find the genetic cause, the team turned to trio-based whole-exome sequencing, analyzing the girl and both parents simultaneously. Sequencing libraries were prepared with Agilent’s SureSelect protocol and run on an Illumina HiSeq 2000, generating 12.45 gigabases of high-quality data with more than 83 percent of target regions covered at twentyfold depth. Filtering for de novo variants present in the child but absent from both parents, and prioritizing genes with relevant clinical associations, the analysis yielded a single compelling candidate: a heterozygous variant in GABBR2, designated c.1699G>A, which substitutes a threonine for an alanine at position 567 of the protein, abbreviated p.Ala567Thr. Sanger sequencing confirmed the finding. It was the only clinically relevant de novo variant in the entire exome.

GABBR2 sits on chromosome 9 at band 9q22.33 and encodes a 941-amino-acid membrane protein belonging to the G protein-coupled receptor family, with seven transmembrane domains. The protein forms a subunit of the GABA-B receptor, the metabotropic receptor for gamma-aminobutyric acid, the central nervous system’s main inhibitory neurotransmitter. Unlike the ion-channel GABA-A and GABA-C receptors, GABA-B receptors modulate neuronal activity by regulating neurotransmitter release, and the receptor is highly expressed across many brain regions. Crucially, prior work has already tied GABAergic dysfunction to Rett biology: mice lacking MeCP2 specifically in GABAergic neurons initially appear normal but then develop forepaw stereotypies, compulsive grooming, impaired motor coordination, abnormal EEG hyperexcitability, severe respiratory dysrhythmias, and premature death. Because MeCP2 is essential for normal GABAergic neuron function, a pathogenic variant in a GABA-B receptor subunit could plausibly converge on the same inhibitory signaling pathways that MeCP2 disruption impairs.

Remarkably, the p.Ala567Thr variant is not a one-off. It affects a highly conserved residue within the third transmembrane domain, and a literature review identified seventeen heterozygous de novo GABBR2 variants in patients with intellectual disability, autism spectrum disorder, and drug-resistant epilepsy, seven of which were this exact variant. With the new case, ten patients with p.Ala567Thr have now been reported, making the Swedish girl the eleventh. The first, described in 2016, was a girl with developmental stagnation at seven months followed by regression, absent language, hand stereotypies, hyperventilation, bruxism, abnormal sleep, crying spells, autistic features, and small feet, though she never had seizures. The same variant appeared in another patient meeting Rett criteria that same year, and in 2017 researchers screening 34 MECP2-negative patients with Rett-like presentations found it in two more, providing functional evidence of reduced receptor activity in both in vitro and in vivo models.

Across these eleven patients, seven fulfill clinical Rett criteria, though most were classified as atypical rather than classic. The most common features are intellectual disability, autism spectrum disorder, and an Rett-like phenotype, while only about half have epilepsy, despite GABBR2’s established links to epileptic encephalopathy. The Swedish girl stands out as one of the most Rett-like cases reported to date, meeting criteria for classic rather than atypical Rett syndrome. OMIM currently lists two phenotype entries for GABBR2, neither formally linked to Rett syndrome, though one, neurodevelopmental disorder with poor language and loss of hand skills, is described as reminiscent of Rett. The overlap across these diagnostic categories suggests that GABBR2-related disease is best understood as a phenotypic spectrum rather than a single discrete syndrome, with the specific variant location and transmembrane domain involved shaping the resulting clinical picture.

The case also carries therapeutic implications. The girl’s self-injurious behavior and sleep problems have been treated with clonidine, benzodiazepines, aripiprazole, sertraline, melatonin, promethazine, and risperidone, with risperidone offering the best but only short-lived effect. Notably, benzodiazepines had no effect on her self-injury, a finding the authors attribute to mechanism: benzodiazepines enhance inhibition through the GABA-A receptor, whereas the variant disrupts the GABA-B receptor. Baclofen, a GABA-B agonist, is the only drug acting on this receptor, but its low blood-brain barrier penetration, short duration, and rapid tolerance limit its usefulness. More promising avenues include positive and negative allosteric modulators of GABA-B receptors and transcranial magnetic stimulation, which is partly mediated through GABA-B-dependent inhibition. The p.Ala567Thr variant reduces receptor activity to roughly 30 percent of wild-type levels, so restoring receptor function could meaningfully benefit affected patients.

The authors conclude that patients with an Rett phenotype may carry pathogenic variants in genes other than MECP2, and that as diagnostic strategies shift toward broader sequencing, the recurrent p.Ala567Thr variant may prove more common than currently recognized. They argue that GABBR2 should be included in testing panels when comprehensive approaches such as whole-genome sequencing are unavailable, and that phenotype-driven variant prioritization can reduce the risk of missing cases. Given the severe, lifelong neurodevelopmental burden of Rett spectrum conditions, and the possibility that future therapies will target GABAergic dysfunction directly, an early molecular diagnosis may become increasingly consequential for families and clinicians alike.

Subject of Research: A de novo GABBR2 GABA-B receptor variant identified as the cause of classic Rett syndrome in an MECP2-negative girl

Article Title: A Heterozygous Variant in the GABBR2 Gene in a Girl With Clinical Classic Rett Syndrome

Article References: Klintenstedt, J., Baeck, P., Engerström, I. W., & Gunnarsson, C. (2026). A Heterozygous Variant in the GABBR2 Gene in a Girl With Clinical Classic Rett Syndrome. Molecular Genetics & Genomic Medicine, 14(10), Article e70303. https://doi.org/10.1002/mgg3.70303

Image Credits: AI Generated

DOI: 10.1002/mgg3.70303

Keywords: Rett syndrome, GABBR2, GABA-B receptor, MECP2, whole-exome sequencing, neurodevelopmental disorder, GABAergic signaling, de novo variant, autism spectrum disorder, intellectual disability, genotype-phenotype correlation, epileptic encephalopathy