Abstract
Original Article Citation 1:
Baldassari S, Klingler E, Teijeiro LG, Doladilhe M, Raoux C, Roig-Puiggros S, Bizzotto S, Couturier J, Gilbert A, Sami L, Ribierre T, Aronica E, Adle-Biassette H, Chipaux M, Jabaudon D, Baulac S. Nat Neurosci. 2025;28(5):964-972. doi:10.1038/s41593-025-01936-z Focal cortical dysplasia type II (FCDII) is a cortical malformation causing refractory epilepsy. FCDII arises from developmental somatic activating mutations in mTOR pathway genes, leading to focal cortical dyslamination and abnormal cytomegalic cells. Which cell types carry pathogenic mutations and how they affect cell-type-specific transcriptional programs remain unknown. In the present study, we combined several single-nucleus genotyping and transcriptomics approaches with spatial resolution in surgical cortical specimens from patients with genetically mosaic FCDII. Mutations were detected in distinct cell types, including glutamatergic neurons and astrocytes, and a small fraction of mutated cells exhibited cytomegalic features. Moreover, we identified cell-type-specific transcriptional dysregulations in both mutated and nonmutated FCDII cells, including synapse- and neurodevelopment-related pathways, that may account for epilepsy and dysregulation of mitochondrial metabolism pathways in cytomegalic cells. Together, these findings reveal cell-autonomous and non-cell-autonomous features of FCDII that may be leveraged for precision medicine. Original Article Citation 2:
Fang C, Zhang X, Yang L, Sun L, Lu Y, Liu Y, Guo J, Wang M, Tan Y, Zhang J, Gao X, Zhu L, Liu G, Ren M, Xiao J, Zhang F, Ma S, Zhao R, Mei X, Qi D. Nat Commun. 2025;16(1):3320. doi:10.1038/s41467-025-58535-6 Focal cortical dysplasia type II (FCDII) is a major cause of drug-resistant epilepsy, but genetic factors explain only some cases, suggesting other mechanisms. In this study, we conduct a molecular analysis of brain lesions and adjacent areas in FCDIIb patients. By analyzing over 217,506 single-nucleus transcriptional profiles from 15 individuals, we find significant changes in smooth muscle cells (SMCs) and astrocytes. We identify abnormal vascular malformations and a unique type of SMC that we call “Firework cells”, which migrate from blood vessels into the brain parenchyma and associate with VIM+ cells. These abnormalities create localized ischemic–hypoxic (I/H) microenvironments, as confirmed by clinical data, further impairing astrocyte function, activating the HIF-1α/mTOR/S6 pathway, and causing neuronal loss. Using zebrafish models, we demonstrate that vascular abnormalities resulting in I/H environments promote seizures. Our results highlight vascular malformations as a factor in FCDIIb pathogenesis, suggesting potential therapeutic avenues.
Commentary
Focal cortical dysplasia type II (FCDII) is a developmental malformation of the human cerebral cortex and among the top causes of drug-resistant epilepsy. The hallmark pathological features of FCDII are disruption of normal cortical layering and the presence of distinctive, infrequent cytomegalic cells—specifically dysmorphic neurons (DNs) and, in type IIb, balloon cells (BCs). Intriguingly, both DNs and BCs frequently harbor identical somatic variants that activate the mammalian target of rapamycin (mTOR) pathway, suggesting their origin from a common neural progenitor. 1 However, variant-carrying cells form only a minority of all cells within the FCDII lesion, and a minority of variant-carrying cells are cytomegalic DNs and BCs. This prompts several central, unresolved questions: Are DNs and BCs fundamentally different cell types or divergent states of the same lineage? Are these rare, visibly abnormal cells by themselves sufficient to cause epilepsy, or are they just the tip of the iceberg? Do the majority reference-sequence (nonmutated) cells and the nonneural microenvironment, including vascular components, play critical roles in cortical dyslamination and epileptogenesis?
Recent advances using single-nucleus RNA sequencing (snRNA-seq) on resected FCDII brain tissue, with validation from immunohistochemistry and other orthogonal methods, have provided new and sometimes conflicting answers.2,3 Baldassari et al. demonstrated that although DNs show evidence of mitochondrial dysfunction, they represent only a small fraction of the population of mTOR-variant-bearing cells. Moreover, cells with the reference (nonmutant) mTOR sequence display altered synaptic gene expression. 2 By contrast, Fang et al. sought to uncover cellular and molecular mechanisms underlying FCDIIb pathogenesis by examining gene expression across diverse cell types regardless of mTOR-variant status, discovering instead an abnormal, novel cell population exhibiting vascular smooth muscle cell (SMC) markers, termed “firework cells.” Their results point to vascular pathology and hypoxia as potential contributors to epileptogenesis in FCDIIb. 3 Together, these studies show that FCDII involves a broad spectrum of aberrant cell types and microenvironmental changes, challenging narrow models focused solely on mutant cytomegalic cells.
In the work by Baldassari et al., tissue from 15 FCDII patients (each with distinct mTOR pathway mutations) and 11 nonepileptic controls was examined using a suite of molecular techniques: snRNA-seq, spatial transcriptomics, and laser-capture microdissection, in combination with immunohistochemical and ultrastructural analysis. They found mTOR pathway mutations widely dispersed across neuronal and glial lineages, not limited to rare cytomegalic cells. The largest transcriptional disturbances relative to controls were seen in mutated glutamatergic neurons and astrocytes. Within FCDII glutamatergic neurons, the presence of an mTOR variant led to upregulated genes involved in mitochondrial metabolism; this was corroborated by immunohistochemistry and electron microscopy showing swollen and damaged mitochondria in DNs. To probe broader network dysfunction, they compared reference-sequence neurons from FCDII to those of control brains, revealing aberrant expression patterns in synaptic signaling genes, a sign of circuit-level disruption within the lesional cortex. Spatial transcriptomics also clarified the distinct molecular identities and distribution of cytomegalic cells: DNs map to glutamatergic neuron populations in gray matter, while BCs display signatures consistent with astrocytes dispersed in both white and gray matter.
The findings of Fang et al. diverge in crucial ways. Comparing lesional to adjacent nonlesional cortex from 15 FCDIIb patients, they used snRNA-seq and immunohistochemistry to identify dysregulation concentrated in astrocytes and SMCs rather than in glutamatergic neurons. Immunostaining for SMC markers ACTA2 and TAGLN identified a novel cell type called “firework cells” (FCs), named for their extensive, branching processes. The abundance of FCs correlated with indicators of FCD pathology, including neuronal loss, elevated vimentin (a marker present in both DNs and BCs), and increased phosphorylated S6 (a well-known indicator of mTOR pathway activation). Vascular staining further revealed abnormally increased, thickened, and sometimes infarcted blood vessels. To test if vascular disruption could increase seizure propensity, they treated zebrafish larvae with a vascular endothelial growth factor (VEGF) receptor inhibitor, a manipulation known to impair normal vascular function, which heightened their susceptibility to pentylenetetrazole-induced seizures. Analysis suggested that FCs express genes tied to cytoskeletal remodeling and migration, which could equip them to dissociate from vessel walls, enter the brain parenchyma, destabilize vasculature, and deplete regular SMCs from arterioles, collectively reducing cerebral perfusion and generating hypoxia, a state known to exacerbate epileptogenesis.
These two landmark studies leverage innovative technology to deliver an unbiased, cell-type-resolved view of FCDII pathology, refining earlier models and providing new “ground truth” from human tissue. Fang et al. heightened the rigor of their findings by using brain tissue adjacent to the lesion as internal controls, minimizing artifact from genetic or procedural differences. Their discovery of a vascular element in FCDIIb aligns with previous work implicating VEGF signaling and blood–brain barrier leakage in related disorders such as tuberous sclerosis, 4 as well as morphological evidence of hypoxia and vascular malformations in other cortical dysplasias.5,6 Notably, although FCs morphologically resemble reactive astrocytes, which can also express ACTA2, 7 their expression of the smooth muscle marker TAGLN and lack of the astrocyte marker glial fibrillary acidic protein support that they are a novel cell type rather than simply reactive glia. The strength of work from Baldassari et al. lies in their integration of spatial, ultrastructural, and single-nucleus transcriptomic data, clarifying that cytomegalic DNs and BCs constitute only a small percentage of all mutated cells, with the majority of lesional tissue composed of reference-sequence cells whose transcriptomes are also perturbed. Many assignments of DN and BC identity to glutamatergic and astrocytic lineages, respectively, had been suggested by previous histological studies, but the single-cell, transcriptome-wide approach now rigorously establishes these designations.
The contrasting results reflect divergent study goals: Baldassari et al. used targeted deep sequencing to quantify mTOR-variant allele frequencies by cell type, whereas Fang et al. focused on transcriptional and cellular phenotypes across FCDIIb lesions regardless of the presence of an underlying genetic variant. Still, most prior studies support that mTOR pathway mutations and hyperactivity are widespread in FCDII, suggesting a primary genetic etiology with possible secondary or parallel contributions from vascular pathology. Both studies are inherently limited to correlations; the directionality of causation, whether mitochondrial or synaptic dysfunction and hypoxia are cause or consequence of recurrent seizures and tissue remodeling, remains speculative. Similarly questionable is the notion that hypoxia chronically activates HIF1α/mTOR/S6 signaling in these lesions, since experimentally, hypoxia typically inhibits mTOR activity. 8 The causal link between FCs and seizure generation is not conclusively demonstrated: the zebrafish model showed enhanced seizures with general VEGF receptor inhibition, but not direct FC manipulation, and inhibition of VEGF signaling can also affect neurogenesis. Additionally, these findings contrast with prior work in tuberous sclerosis, where inhibiting VEGF ameliorates, rather than worsens, pathology. 4
These insights open important new questions. The observation that mTOR pathway mutations are present in microglia, endothelial cells, interneurons, and other diverse cell types raises the question of developmental timing: how can somatic mutations arise early enough to populate different lineages, yet remain relatively circumscribed spatially? What are the functional consequences of mTOR mutations in nonneuronal cells such as microglia? When, during development, do DNs and BCs bifurcate into their respective lineages, and how do timing and context of somatic mutation impact lesion properties and seizure risk? The developmental triggers for FC emergence and their specificity to FCDIIb compared to other focal epilepsies, such as mesial temporal sclerosis, also warrant further exploration.
These studies also highlight new therapeutic directions. Mitochondrial dysfunction in DNs suggests a potential benefit from interventions enhancing oxidative phosphorylation or mitophagy. 9 Addressing vascular dysregulation by repairing the blood–brain barrier or improving cortical perfusion may mitigate hypoxia-driven epileptogenesis, though these strategies require translation and testing in experimental systems such as animal models or vasculature-containing cerebral organoids that recapitulate the developmental and neurophysiological features of FCDII.
Ultimately, these 2 studies illuminate FCDII as a disorder with multiple layers of complexity, involving interdependent neuronal, glial, and vascular elements within a malfunctioning cortical network. The cytomegalic DNs and BCs are only the “tip of the iceberg,” while largely hidden beneath are significant, lineage-spanning abnormalities and environmental disruptions that may influence epileptogenesis. The road forward in research and clinical care may require targeting not just rare mTOR-variant cells but also the broader reference-sequence cell population and the cerebral vasculature, integrating genetic, cellular, and microenvironmental insights into comprehensive therapeutic strategies.
