The condition arises from the heterozygous deletion of a 1.5 to 1.8 megabase region on chromosome 7q11.23, a stretch of DNA containing 26 to 28 genes. Occurring in roughly one in 7,500 live births, the deletion accounts for a strikingly complex clinical picture. Children with the classic 1.5 megabase deletion, which represents about 95 percent of cases, show a distinctive paradox: an intense drive toward social contact, often described as hypersociability, coupled with poor awareness of personal boundaries, social disinhibition, and difficulty sustaining friendships. Alongside this, they frequently experience attention deficit hyperactivity disorder, anxiety disorders, specific phobias, and heightened sensitivity to sound. Cognitive impairment is typically mild to moderate, with intelligence quotients averaging between 50 and 60, and visuospatial abilities are disproportionately affected, while fine motor skills are also commonly impaired.

Neuroimaging studies have begun to explain how this behavioral profile emerges from altered brain structure. Magnetic resonance imaging reveals total brain volume reductions of 10 to 15 percent in affected individuals, accompanied by diminished gray matter in the thalamus and the parieto-occipital regions that underpin the human visual-spatial system. Functional abnormalities in the hippocampal formation contribute to deficits in spatial navigation and long-term memory, while alterations in limbic circuitry, particularly the amygdala-prefrontal circuitry, are closely linked to hypersociability and increased anxiety. Intriguingly, cerebellar volumes relative to intracranial volume are increased, a finding that may relate to the motor deficits and poor short-term memory seen in patients.

The central challenge for researchers has been translating these clinical observations into testable biology. Because clinical cohorts are small, variable, and constrained by ethical considerations, animal and organoid models have become indispensable. The review, led by Ya-Yue Chen and colleagues at Zhejiang University School of Medicine, synthesizes two decades of progress across these platforms, with particular emphasis on how they illuminate neurodevelopmental impairments.

Mouse models have proven especially valuable because the genes in the Williams syndrome critical region are conserved between humans and mice, clustered on mouse chromosome 5G2. In 2009, Li and colleagues generated proximal and distal deletion strains using Cre-loxP technology, which when crossed produce double heterozygous mice mimicking the human deletion. These mice exhibit reduced brain volumes, altered neuronal distribution in the somatosensory cortex, heightened social interest, and impaired motor coordination. A landmark advance came in 2014, when the laboratory of Victoria Campuzano created a complete deletion model spanning Gtf2i to Fkbp6, avoiding the complicating homozygous loss of Limk1 present in earlier strains and more accurately replicating the human condition.

Complete deletion mice have since become the workhorse of the field. They display increased startle responses to acoustic stimuli, impaired fear memory, and hypersociability, mirroring the human phenotype. Studies of infant and adolescent mice have revealed reduced body growth, delayed sensory development, and altered patterns of ultrasonic vocalizations. Perhaps most importantly, these mice have become a platform for preclinical drug screening. Combined oral administration of verapamil, an L-type calcium channel blocker, and curcumin for five weeks significantly improved neurodevelopmental impairments in adolescent mice, apparently by modulating inflammasome-related, MAPK, and PI3K/AKT signaling pathways and reducing the number of activated microglia. Notably, neither drug worked alone. In a separate line of work, daily injections of JZL184, a selective monoacylglycerol lipase inhibitor, specifically normalized the social and cognitive phenotypes of mutant mice, suggesting that modulation of the endocannabinoid system represents a promising therapeutic avenue. By contrast, oxytocin, sometimes proposed as a treatment for social deficits, failed to attenuate fear memory impairments in these animals.

The review also dissects the contributions of individual genes within the deleted region, work that has been guided by patients carrying smaller, atypical deletions. When GTF2I and GTF2IRD1 are retained, visuospatial abilities and intellectual function are notably better preserved, and the intraparietal sulcus, a key node for visuospatial processing, remains structurally unaffected. GTF2I, which encodes a multifunctional transcription factor involved in embryonic development, cell cycle regulation, and epigenetic control, appears to account for roughly 10 to 20 percent of the transcriptional dysregulation seen in patient-derived cells, likely through disrupted interactions with the LSD1 repressive chromatin complex. In 2024, Adams and colleagues used CRISPR/Cas9 to generate GTF2I-deficient human cells and found increased proliferation in neural progenitor cells alongside increased cell death and synaptic dysregulation in resulting neurons and cortical organoids. In mice, heterozygous loss of Gtf2i produces hypersociability, reduced motor coordination, increased anxiety, and auditory hypersensitivity, and conditional deletion in forebrain excitatory neurons causes myelination defects that can be pharmacologically rescued by remyelinating drugs such as clemastine and 4-aminopyridine.

Other genes add further layers of complexity. LIMK1, which regulates the actin cytoskeleton by phosphorylating cofilin, has been implicated in impaired visuospatial constructive cognition, and mouse studies reveal effects on dendritic spines, hippocampal long-term potentiation, and long-term memory that appear to operate through the transcription factor CREB rather than cofilin itself. CLIP2, a microtubule-binding protein, contributes to hippocampus-dependent memory and motor coordination when haploinsufficient. Additional genes, including Ncf1, Eif4h, Stx1a/b, Dnajc30, Fzd9, Nsun5, Hip1, and Cldn3, each produce distinctive phenotypes ranging from mitochondrial dysfunction to impaired oligodendrocyte proliferation and blood-cerebrospinal fluid barrier disruption. Yet the picture is not uniform: Fkbp6, for instance, produces no phenotype unless both alleles are knocked out, a reminder that some symptoms likely emerge from interactions among multiple genes or from physiological differences between mice and humans.

Human-specific insights have come from induced pluripotent stem cell and organoid technologies, which preserve the pathogenic mutations of individual patients while recapitulating molecular signaling pathways of human neurodevelopment. Early work showed that cortical neurons derived from patient cells exhibit prolonged action potential repolarization and deficits in voltage-activated potassium currents, alongside morphological alterations consistent with postmortem findings. More recently, forebrain organoids derived from patients have revealed abnormal proliferation and differentiation of neural progenitor cells and aberrant expression of neurodevelopmental genes. A particularly elegant strategy employs CRISPR/Cas9 editing to create isogenic cell lines that differ only at the 7q11.23 region, allowing researchers to study dosage effects against an identical genetic background. Using this approach in 2024, researchers demonstrated that ribosome biogenesis plays a key role in the neurodevelopmental disorder, a finding with potential therapeutic implications. Gene-edited organoids carrying deletions spanning NSUN5 to GTF2IRD2 have similarly revealed significant downregulation of synaptic genes and pathways relevant to GABAergic neurons.

The authors are candid about limitations on both sides of the modeling divide. The Williams syndrome critical region is inverted between human chromosome 7q11.23 and mouse chromosome 5G2, and broader differences in brain structure and developmental trajectories limit translation; cerebellar volumes, for example, are enlarged in patients but unaffected in some mouse models. Murine microglia differ substantially from their human counterparts, complicating the study of neuroinflammation. Organoids, meanwhile, lack vascularization and immune components, which restricts their use in modeling cardiovascular phenotypes and can lead to hypoxia and limited tissue maturation in long-term cultures. The rarity of clinical samples and the reliance on single-sex iPSC lines in some studies further constrain generalizability.

Future directions outlined in the review include coculturing brain organoids with blood vessel organoids to form neural-specific vascular networks, introducing microglia or microglial progenitors to build immune-competent models, and deploying microfluidic organ-on-a-chip systems that mimic the blood-brain barrier and fluid shear stress. The authors also propose expanding beyond mice to rat, domestic dog, and non-human primate models. Domestic dogs are particularly intriguing because their hypersociability relative to wolves likely stems from structural variations in GTF2I and GTF2IRD1, the very genes implicated in Williams syndrome. Non-human primates offer comparable cognitive skills and social complexity that would allow assessment of fine motor deficits, rhythmic abilities, and intelligence quotients that are difficult to measure in rodents. Emerging technologies, including single-cell and spatial multiomics and optogenetics, are expected to further enhance model fidelity, while advanced neuroimaging in animal models could bridge preclinical findings to clinically useful biomarkers.

The overarching message is one of cautious optimism. For a disorder in which current treatment remains limited to symptomatic management with antidepressants, anxiolytics, stimulants, surgery for vascular anomalies, and educational support, the convergence of complete deletion mice, human forebrain organoids, and precision gene editing is steadily converting descriptive observation into mechanistic understanding. The path forward, the authors argue, lies in integrating these complementary systems into clinical trials designed specifically for Williams syndrome, using biomarkers to stratify patients and developmentally appropriate endpoints to measure success. Such an approach holds promise for moving beyond symptom management toward treatments that target the biological roots of the disorder.

The cerebral cortex, responsible for advanced cognitive functions, contains a diverse repertoire of projection neurons (PNs) that extend axons over long distances to establish functional brain circuits. Among these, interhemispheric-callosal and corticothalamic neurons are two prominent subtypes exhibiting distinct connectivity patterns. The current investigation probes the enigmatic question of how these distinct neuronal populations regulate RNA populations within their growth cones to support subtype-specific functions during neurodevelopment.

By leveraging transcriptomic profiling techniques tailored to isolate RNA from growth cones at multiple developmental time points, the researchers systematically compared the subcellular transcriptomes of callosal versus corticothalamic projection neurons. These analyses unveiled a dual landscape of RNA regulation: a core set of RNAs localized in growth cones shared between both subtypes, and a substantial fraction of subtype-enriched transcripts. This division highlights a balance between conserved molecular machinery essential for general axonal growth and specialized RNA localization that fine-tunes subtype-specific circuit wiring.

The temporal dimension added remarkable insight, revealing dynamic shifts in growth cone-localized transcriptomes corresponding to distinct developmental milestones such as initial axon extension, target innervation, and synapse formation. This temporal regulation suggests that neurons deploy context-dependent RNA localization programs within their growth cones, enabling them to rapidly respond to environmental cues with local protein synthesis tailored to current developmental needs.

Further dissecting the molecular underpinnings, the study focused on sequence elements within the 3′ untranslated regions (3’UTRs) of mRNAs enriched in growth cones, identifying motifs associated with subtype-specific mRNA localization and stability. These sequence elements are hypothesized to serve as docking sites for RNA-binding proteins (RBPs), orchestrating the targeted transport and localized translation of mRNAs critical for axonal development and branching.

Among the RBPs spotlighted, CPEB4 emerged as a key translational regulator selectively influencing the branching complexity of axons in developing neurons. CPEB4’s modulation of polyadenylation and local translation in growth cones delineates a crucial node in the molecular nexus that governs axonal arborization patterns essential for circuit precision. Furthermore, the RNA-binding motif protein RBMS1 was found to play a dynamic, subtype-specific role in shaping callosal neuron circuits, underscoring the nuanced regulation carried out by distinct RBPs in neurodevelopment.

Importantly, the transcriptomic landscapes of these growth cones enriched genes implicated in neurodevelopmental and neuropsychiatric disorders, offering a mechanistic window into how dysregulation of subcellular RNA localization might contribute to brain pathologies. This aligns with growing evidence linking aberrant RNA metabolism with disorders such as autism spectrum disorder and schizophrenia, situating the present findings at a crossroads of basic and translational neuroscience.

The methodologies employed—combining high-resolution subcellular RNA sequencing, motif discovery, and functional perturbations in vivo—set a new standard for dissecting RNA regulation within defined neuronal compartments. This approach offers a framework applicable to other polarized cell types beyond neurons, where spatial control of RNA fate dictates cell behavior and function.

This study shines a spotlight on the growth cone as a critical hub of post-transcriptional gene regulation, emphasizing that neurons do not merely convey genetic information linearly but finely tailor RNA distribution and translation at the subcellular level to choreograph complex developmental programs. Such nuanced RNA regulatory landscapes within neuronal subcellular domains have remained largely unexplored until now.

The implications extend to understanding how cellular polarity and identity are maintained through localized RNA regulation, casting new light on fundamental principles of cell biology. Growth cones exemplify a specialized polarized compartment where local RNA regulation communicates developmental context and subtype identity into precise structural and functional outcomes.

By detailing both conserved and subtype-specific RNA localization programs, the research reveals a molecular grammar governing the developmental trajectory of projection neurons. This grammar, encoded within RNA sequences and interpreted by RNA-binding proteins, acts as a versatile toolkit for neurons to dynamically adapt their growth and target interactions.

Moreover, this work raises intriguing questions about the extent to which RNA localization and translation in growth cones contribute to synaptic specificity and plasticity in mature circuits. It paves the way for future investigations into how localized RNA regulation integrates with extrinsic signals during critical periods of brain wiring.

The broader significance touches upon emerging paradigms where RNA metabolism is key to neural circuit formation, maintenance, and remodeling. Understanding these processes at the molecular level offers promising avenues for therapeutic strategies aimed at restoring or modulating neural connectivity in developmental disorders.

As the study demonstrates, unraveling the complex interplay between RNA sequences, binding proteins, and neuronal subtype contexts enriches our comprehension of brain development’s molecular machinery. It highlights subcellular RNA dynamics as a previously underappreciated layer of gene regulation essential for the nervous system’s structural and functional refinement.

In sum, the revelation that neuronal subtypes deploy distinct, dynamic RNA localization and translation programs in their growth cones transforms our conceptualization of neurodevelopmental gene regulation. It underscores a sophisticated molecular logic where local protein synthesis is tailored not only to developmental stage but also to the specific identity and function of projection neuron subtypes.

This landmark contribution enriches the lexicon of neurobiology, illustrating how intricate RNA regulation contributes to the exquisite cellular diversity and circuit complexity of the cerebral cortex. It opens novel investigative pathways that bridge molecular neuroscience, developmental biology, and neuropathology—an inspiring synthesis poised to accelerate discoveries in brain health and disease.

Subject of Research:
Subcellular RNA localization and regulation within growth cones of distinct neuronal subtypes during cerebral cortex development.

Article Title:
Dynamic subtype- and context-specific subcellular RNA regulation in growth cones of developing neurons of the cerebral cortex.

Article References:
Veeraraghavan, P., Engmann, A.K., Hatch, J.J. et al. Dynamic subtype- and context-specific subcellular RNA regulation in growth cones of developing neurons of the cerebral cortex. Nat Neurosci (2025). https://doi.org/10.1038/s41593-025-02173-0

Image Credits:
AI Generated

DOI:
https://doi.org/10.1038/s41593-025-02173-0