Microtubules are critical structural and functional components of the neuronal cytoskeleton, orchestrating a wide array of processes essential for brain development. These dynamic polymers are composed of α- and β-tubulin heterodimers, which assemble into highly organized arrays that support neurogenesis, neuronal migration, differentiation, and the formation of synaptic architecture and connectivity (Janke and Magiera, 2020; Breuss et al., 2017). The functional diversity of microtubules is further modulated by the differential expression of tubulin isotypes and a range of post-translational modifications, which together fine-tune neuronal migration and intracellular dynamics, leading to the correct development of the brain (Moutin et al., 2021; Wloga et al., 2017; Tantry and Santhakumar, 2023). Proper microtubule function underpins essential developmental processes, including the proliferation of neural progenitors, guidance of immature neurons from periventricular zones to the cortical plate, stabilization of axonal projections, and the formation of synaptic networks (Tantry and Santhakumar, 2023; Kumar et al., 2010; Maillard et al., 2023). To date, at least ten α-tubulin isotypes (i.e., TUBA1A, TUBA1B, TUBA1C, TUBA4A, TUBA4B, TUBA3C, TUBA3D, TUBA3E, TUBA8, and TUBAL3) and seven β-tubulin isotypes (i.e., TUBB, TUBB2A, TUBB2B, TUBB3, TUBB4A, TUBB4B, and TUBB6) have been identified (Gonçalves et al., 2018; Park et al., 2021). Their interactions with key microtubule-associated proteins—including DCX, LIS1, DYNC1H1, FLNA, and Tau—are indispensable for cytoskeletal integrity and neuronal polarity.

Neurodevelopmental disorders (NDDs) comprise a heterogeneous group of frequently overlapping conditions, including developmental delay (DD), intellectual disability (ID), speech and motor impairments, behavioral dysregulation, and autism spectrum disorder (ASD) (Salpietro et al., 2019; de Masfrand et al., 2024; Kotchetkov et al., 2023). Despite the advent of next-generation sequencing (NGS), the molecular underpinnings of NDDs remain unresolved in a substantial proportion of patients (Salpietro et al., 2018; Iacomino et al., 2024). A deeper understanding of the genetic architecture and disrupted biological pathways is essential to improve diagnostic precision and inform targeted interventions.

In recent years, pathogenic variants in tubulin-encoding genes have emerged as causative factors in a broad spectrum of neurodevelopmental and neurological disorders. Given their fundamental roles in neuroanatomical organization, tubulin gene mutations have been strongly linked to malformations of cortical development (MCDs), defining a growing group of disorders collectively termed tubulinopathies (Severino et al., 2020; Bahi-Buisson et al., 2014; Romaniello et al., 2018). The resultant phenotypes reflect diverse pathogenic mechanisms, including impaired neuroblast proliferation, aberrant neuronal migration, defective axonal pathfinding, and disrupted synaptogenesis (Breuss et al., 2017; Bahi-Buisson et al., 2014; Oegema et al., 2015; Breuss et al., 2017). Clinically, tubulinopathies encompass a broad neurodevelopmental spectrum including microcephaly, variable neurodevelopmental impairment including motor delay, intellectual disability, speech difficulties and broad behavior abnormalities, as well as frequent movement disorders, stereotypies and epilepsy (Brock et al., 2021; Schmidt et al., 2021). Neuroimaging typically reveals characteristic abnormalities, such as polymicrogyria, lissencephaly-pachygyria, dysgyria, agenesis of the corpus callosum, basal ganglia dysmorphism, brainstem asymmetry, and cerebellar hypoplasia or dysplasia (Breuss et al., 2017; Patel et al., 2017; Cushion et al., 2014; Ejaz et al., 2017).

While variants in TUBA1A, TUBB2B, and TUBB3 account for the majority (>90%) of known tubulinopathy-associated mutations (Gonçalves et al., 2018; Bahi-Buisson et al., 2014), Pathogenic variants in other tubulin genes (including TUBB2A) are considerably rarer and less well characterized. Although initial reports suggested stereotyped phenotypes for mutations in specific tubulin genes, subsequent studies have challenged this notion, revealing marked inter- and intra-genic variability (Bahi-Buisson et al., 2014; Amrom et al., 2014; Zanni et al., 2013; Guerrini et al., 2012; Cushion et al., 2013).

TUBB2A encodes a β-tubulin isotype predominantly expressed in the developing brain, although its relative expression appears lower compared to other isoforms (Breuss et al., 2012). Pathogenic variants in TUBB2A are extremely rare and currently account for only a small fraction of tubulinopathy cases (Schmidt et al., 2021). To date, only 24 individuals have been reported with TUBB2A-related disease, showing wide variability in both neurodevelopmental trajectories and neuroimaging findings, ranging from near-normal cortical architecture to severe cortical malformations (Cushion et al., 2014; Ejaz et al., 2017; Brock et al., 2021; Schmidt et al., 2021; Cai et al., 2020; Sferra et al., 2018; Rodan et al., 2017; Lee et al., 2014).

Here, we expand the phenotypic and mutational spectrum of TUBB2A-associated tubulinopathy by presenting four additional individuals harboring a novel heterozygous missense variant (NM_001069.3:c.1172G > A; NP_001060.1:p.Arg391His). This variant affects a highly conserved residue in the C-terminal domain of β-tubulin and highlights a mutational hotspot with significant implications for microtubule dynamics and α/β-tubulin heterodimer formation, leading to disorders in brain development and function.

The assembly of α- and β-tubulin isotypes into microtubules is a fundamental process underlying neuronal development and connectivity. These microtubule networks mediate diverse cellular events, including neurogenesis, neuronal migration, axonal guidance, and the formation of both cortical and subcortical brain architecture (Bahi-Buisson et al., 2014; Cushion et al., 2013; Breuss et al., 2012). Perturbations in tubulin dynamics are increasingly recognized as the molecular basis of a broad and clinically heterogeneous group of neurodevelopmental disorders termed tubulinopathies (Bahi-Buisson et al., 2014; Romaniello et al., 2018). These disorders arise from heterozygous pathogenic variants in genes encoding α- or β-tubulins, and are commonly associated with a spectrum of malformations of cortical development (MCD), as well as anomalies affecting subcortical structures such as the basal ganglia, corpus callosum, and cerebellum (Bahi-Buisson et al., 2014; Cushion et al., 2014; Cushion et al., 2013; Brock et al., 2021; Schmidt et al., 2021).

In this study, we describe four individuals carrying a novel recurrent TUBB2A missense variant (NM_001069.3:c.1172G > A; NP_001060.1:p.Arg391His), expanding the mutational and phenotypic spectrum associated with TUBB2A-related tubulinopathy. Importantly, the recurrence of the p.Arg391His variant across unrelated families raises the possibility of a mutational hotspot or, alternatively, low-level germline mosaicism, which may underlie its repeated occurrence.

Interestingly, all subjects presented with microcephaly and a simplified gyral pattern at the time of presentation. Temporo-basal dysgyria was observed in two individuals, and cerebellar foliar anomalies were noted in three cases. The corpus callosum appeared disproportionately large relative to the degree of microcephaly. Notably, none of the patients exhibited the hallmark features of tubulinopathy, such as asymmetric ventriculomegaly and basal ganglia dysmorphisms. Moreover, dysgyria was noted only in two cases and was limited to the temporo-basal regions. One individual demonstrated a progressive phenotype, developing cerebellar atrophy over an eight-year period. Cerebellar atrophy is, in fact, a recognized feature in certain tubulinopathies, particularly those linked to mutations in the TUBB4A gene, such as Hypomyelination with Atrophy of the Basal Ganglia and Cerebellum.

We systematically compared these individuals with the 24 previously reported cases in the literature, resulting in a combined cohort of 28 patients (Table 1) (Cushion et al., 2013; Brock et al., 2021; Schmidt et al., 2021; Cai et al., 2020; Sferra et al., 2018; Rodan et al., 2017; Lee et al., 2014). Analysis of this cohort reveals a highly penetrant core phenotype characterized by neurodevelopmental delay or intellectual disability in all individuals (100%), predominantly of moderate-to-severe severity (89.2%, 25/28). Language impairment was universally present (100%), with complete absence of speech in 64.2% (18/28). Autism spectrum disorder (ASD) or ASD traits were noted in 39.2% (11/28), reinforcing the neuropsychiatric vulnerability associated with TUBB2A dysfunction.

Postnatal microcephaly was present in nearly half of the patients (46.4%, 13/28), underscoring the role of TUBB2A in early neurogenesis and brain volume regulation. Epilepsy was a prominent comorbidity, affecting 78.5% (22/28), with 63.6% of cases manifesting seizures before 12 months of age. Seizure onset varied widely, ranging from infancy to mid-childhood. Motor deficits were reported in 82.1% (23/28), including axial hypotonia (35.7%), spastic diplegia or paraplegia (7.1%), and marked delays in achieving motor milestones (75%). Dysmorphic facial features were present in 28.5% (8/28), and visual impairment in 14.2% (4/28), though these features were not uniformly observed. Neuroimaging findings further support the pathogenic role of TUBB2A variants. MRI abnormalities were present in all individuals, with 67.8% (19/28) exhibiting cortical malformations. Pachygyria (21.4%) and diffuse dysgyria (32.1%) were the most common cortical anomalies, while dysgenesis or thinning of the corpus callosum was observed in 64.2% (18/28). Additional structural changes included cerebellar vermis or hemispheric hypoplasia (21.4%), basal ganglia dysplasia (21.4%), white matter abnormalities (25%), and ventricular enlargement (10.7%). These imaging findings emphasize the widespread disruption of both cortical and subcortical development resulting from TUBB2A dysfunction.

Among our newly described individuals, two (Individuals 1 and 2) are siblings born to unaffected, non-consanguineous parents. Sanger sequencing confirmed the absence of the variant in both parents, indicating germline mosaicism as the likely mechanism of transmission although additional tests were nor performed based on parents’ decision. This is the first documented case of TUBB2A-related tubulinopathy due to gonadal mosaicism, contrasting with all previously published TUBB2A variants, which have arisen de novo. This finding underscores the importance of considering recurrence risk in familial settings, even when initial testing suggests de novo origin. The identified variant, p.Arg391His, affects a highly conserved residue within the C-terminal domain of β-tubulin, localized at the longitudinal E-site heterodimer interface—a critical surface for α/β-tubulin dimerization (Figure 1E). To assess the structural and energetic consequences of this mutation, we employed high-resolution molecular modeling based on cryo-electron microscopy structures of human α/β-tubulin heterodimers (PDB IDs: 8ixa, 8ixd, 8ixf for TUBA1A-TUBB2A, TUBA1C-TUBB2A, and TUBA4A-TUBB2A, respectively). In all three models, Arg391 of β-tubulin forms conserved electrostatic and hydrophobic interactions with Tyr262, Trp346, and Glu434 of the α-tubulin subunit (Figures 2A–2D), stabilizing the heterodimer interface. The introduction of the Arg391His substitution was found to disrupt this conserved interaction network. The imidazole side chain of histidine is unable to replicate the native arginine’s interaction profile, resulting in altered side-chain packing and reduced electrostatic complementarity. Binding free energy calculations revealed a consistent energetic penalty across all three heterodimer models, with higher binding free energies in mutant complexes, indicating weakened dimerization (Figure 2E). Further analysis of the electrostatic surface potential revealed loss of electrostatic complementarity in all mutant complexes, with reduced Pearson correlation coefficients relative to wild-type pairs (Figure 2F). While solvation energies were slightly more favorable in mutant dimers (data not shown), this minor gain did not compensate for the loss of stabilizing electrostatic interactions. These findings strongly support a destabilizing effect of the Arg391His mutation on α/β-tubulin heterodimer formation, which may underlie the observed neurodevelopmental phenotype by compromising microtubule integrity and dynamics during early brain development (Soler et al., 2025).

Taken together, the convergence of clinical, neuroimaging, genetic, and computational data reinforces the role of Arg391 as a mutational hotspot in TUBB2A, with key structural and functional implications for microtubule assembly.

This study provides a comprehensive clinical and molecular characterization of TUBB2A-related neurodevelopmental disorders, incorporating four newly described patients—including the first familial recurrence due to gonadal mosaicism—and a detailed comparison with previously reported cases. The identification of a novel pathogenic variant, p.Arg391His, at a highly conserved and structurally critical interface, provides important insights into genotype–phenotype relationships and disease mechanisms. Through integrative structural modeling, we demonstrate that Arg391 plays a key role in maintaining heterodimer stability across multiple α-tubulin isoforms. Its substitution by histidine disrupts a conserved binding network, weakening α/β-tubulin dimerization and, by extension, microtubule function. These findings add to growing evidence that TUBB2A mutations can produce a spectrum of tubulinopathy phenotypes through impaired microtubule assembly.

However, our study has certain limitations. The lack of functional in vivo evidence—such as experiments using cultured cells or patient-derived neuronal models—restricts our ability to fully evaluate the physiological relevance of the p.Arg391His variant in microtubule dynamics. Notably, Luscan and colleagues conducted insightful in vivo studies on the same residue (c.1172G > A [p.Arg391His]; c.1171C > T [p.Arg391Cys]) in TUBB4B, employing COS-7 cell transfection and patient fibroblast assays. Their findings demonstrated deleterious effects of these substitutions on microtubule repolymerization dynamics and growth rates (Luscan et al., 2017). Given the high structural homology among tubulin subunit genes, these results provide valuable context for interpreting our observations.

Future analogous studies should aim to elucidate the impact of TUBB2A mutations in neuronal models and to investigate whether therapeutic stabilization of dimer interfaces might offer a viable intervention strategy. The recurrent involvement of Arg391 positions this site as a candidate target for future precision medicine approaches, with implications for both diagnosis and therapy in tubulinopathy-related neurodevelopmental disorders.