Gene Variants Behind Periventricular Nodular Heterotopia Epilepsy

Periventricular nodular heterotopia, or PVNH, arises when clusters of neurons fail to reach the cortex and instead pile up in nodules along the walls of the brain’s ventricles. A review published in Acta Epileptologica now brings together more than two decades of clinical reports into an updated spectrum of 23 risk genes whose likely pathogenic variants lead to PVNH in human patients. The authors, Jianping Song, Xiaoqin Sun and Chunqing Zhang, organize the genes chronologically by when the first patient was reported and describe, for each one, the physiological function of its protein product and the mechanisms by which single genetic changes can derail brain development.

PVNH belongs to a broader family of disorders called malformations of cortical development, which include gray matter heterotopia caused by blocked neuronal migration. In PVNH, the integrity of the ventricular lining is broken, and neurons gather in nodules containing pyramidal cells with immature morphology and haphazardly oriented apical dendrites, along with distinct subpopulations of GABAergic interneurons. The nodules most often involve the trigone and occipital horn of the lateral ventricles, followed by the frontal horn and body, and are frequently accompanied by ventriculomegaly and hydrocephalus. Based on their distribution, the review distinguishes five types of nodules: bilateral and symmetrical, bilateral single-noduled, bilateral and asymmetrical, unilateral, and unilateral with extension into the neocortex.

Clinically, PVNH is one of the major causes of hereditary epilepsy, accounting for 13 to 20 percent of patients who have both a malformation of cortical development and seizures. Seizures can begin as early as a few days after birth, though they more commonly appear between 10 and 20 years of age, and generalized tonic–clonic seizures are frequently observed. The main clinical triad consists of neurological dysfunction, intellectual disability and refractory epilepsy. Stereo-electroencephalogram recordings show that most seizures originate simultaneously from the nodules and the functionally related cortex, while others arise from the heterotopias alone, and fiber tractography imaging has revealed abnormal nerve fiber connections between the ectopic nodules and the overlying cortex.

The gene most closely associated with the condition is FLNA on the X chromosome, first linked to PVNH in 1998. Because FLNA-related PVNH follows an X-linked dominant inheritance pattern, it occurs more commonly in female familial cases, since most male variant carriers do not survive in utero or through the first years of life. Affected individuals may show intellectual disability, developmental delay, focal seizures and extracranial complications such as cardiac valvular dysplasia. FLNA encodes filamin A, an actin-binding protein that crosslinks actin filaments and regulates remodeling of the actin cytoskeleton, a process critical for cell shape and migration. In Flna knockdown rats, the neuroepithelial lining of the brain is perturbed, likely because the apical complex integrity and polarized scaffolds of radial glial cells become disorganized, while Flna knockout male mice die at birth with severe vascular and cardiac defects comparable to the pattern seen in human patients.

Other genes act through entirely different cellular machinery. ARFGEF2, inherited in an autosomal recessive pattern, has been reported in 15 cases of PVNH with microcephaly, epilepsy, quadriplegia and developmental delay, sometimes with dystonia and recurrent respiratory infections. Its product, BIG2, guides intracellular vesicle trafficking together with ARF1, and inhibiting BIG2 in mice disrupts the transport of cell adhesion molecules such as E-cadherin and β-catenin from the Golgi apparatus to the cell membrane, potentially impairing neuroependymal integrity, radial glial scaffolding and glial-guided neuronal migration. A physical interaction between BIG2 and filamin A is reflected in Arfgef2 knockdown mice, which show overexpression of phosphorylated FLNA at serine 2152 and weakened FLNA-actin binding alongside PVNH. Genes such as LRP2, ARX, DCHS1 and FAT4, SHOC2, ERMARD, SCN1A, AKT3, ARF1, NEDD4L, GNAI2, INTS8, TMTC3, MOB2, KLHL7, MAP1B, PLEKHG6, KCNT1, MEN1, ECE2, LAMC3 and SYNCRIP each add further mechanisms, from disrupted cytoskeleton organization and cell adhesion to nucleus-cilia coupling and ion channel dysfunction.

Several of these genes illustrate how varied the resulting disease can be. In Van Maldergem syndrome, biallelic DCHS1 or FAT4 mutations were found in seven of nine reported patients, including individuals and sibling pairs with PVNH ranging from bilateral nodular or laminar subependymal heterotopia to double cortex appearances; the encoded proteins form a receptor-ligand pair that builds an apically located adhesive complex anchoring radial glial cells in the developing brain. Three patients with SCN1A variants showed bilateral PVNH together with electroclinical features consistent with the Dravet syndrome spectrum, leading the authors to recommend brain MRI assessment when SCN1A-associated epilepsy is suspected. A proband with a MEN1 mutation showed bilateral posterior PVNH, refractory focal epilepsy and morbid obesity without the multiple endocrine neoplasia type 1 syndrome; the variant was detected only in brain tissue collected through SEEG electrodes, and the authors note that quantitative identification of MEN1 copy number variation is needed because mosaicism could not be fully characterized.

Animal and organoid models provide much of the mechanistic evidence, though with important caveats. Rodents are the most commonly used models, and most studies have emphasized direct abnormalities of neuronal migration in genes including FLNA, DCHS1, FAT4, GNAI2, MOB2, MAP1B, PLEKHG6 and ECE2, tied to radial glial cell morphology, the cytoskeleton, adhesion, apical anchoring and nucleus-cilia coupling. Yet differences in species evolution make it difficult to map these findings onto human brain development: a well-matched PVNH phenotype is hard to produce in animal models with genetic intervention, and Plekhg6 knockdown mice even develop cobblestone lissencephaly, a malformation involving excessive neuronal migration, the opposite pathology. Complete knockout of Arfgef2, Arf1, Lrp2 or Men1 causes embryonic lethality, likely reflecting differences in mouse basal radial glial cells compared with the highly proliferative and abundant human equivalents, as well as human gene expression complexity in the cortical germinal zones.

Human cerebral organoids are increasingly filling this gap. Organoids derived from induced pluripotent stem cells of patients carrying FAT4 or DCHS1 variants, or from isogenic knockout lines, replicate PVNH, showing defective radial glial morphology and altered migratory dynamics in a subset of neurons. Increased expression of the primate-specific isoform of Plekhg6 in organoids produces accumulation of ectopic neurons at the ventricular surface, a defective apical adherent junction belt and alterations of the radial glial scaffold, while ECE2 modulation organoids avoid confounding variables introduced in animal models and align more closely with clinical manifestations. The review also highlights recent advances in activity-dependent gene therapy: a cell-autonomous strategy using the c-Fos promoter to drive Kv1.1 potassium channel expression selectively modulated hyperactive neurons in epilepsy models without disrupting normal function, offering a framework that could eventually be adapted to target pathogenic neuronal activity in PVNH-associated genetic defects, although targeted therapies for PVNH itself remain lacking.

The authors conclude that while FLNA has been well studied, PVNH caused by other single-gene variants remains poorly understood in clinical practice given the rarity and sporadic nature of the diagnosis. Nodule distribution may vary according to the mutated gene, but all patients face complicated symptoms, and intractable epilepsy, though most commonly observed, is not always present. Because neuronal migration is most active between 12 and 24 weeks of gestation, ultrasound and MRI during pregnancy can potentially identify PVNH, an approach already achieved in a fetus with LAMC3 variants whose parents chose to terminate the pregnancy at 21 weeks and 5 days of gestation based on ultrasound results. Seeking the genetic background of PVNH through exploratory mutation identification in cohorts, the authors argue, is more helpful for early intervention and provides directions for further basic research, and they recommend that patients with PVNH complete exome analysis for more appropriate treatment recommendations and reassurance in future pregnancies.

Subject of Research: Genetic variants that cause periventricular nodular heterotopia, a malformation of cortical development linked to epilepsy

Article Title: Gene variants in periventricular nodular heterotopia

Article References: Song, J., Sun, X., & Zhang, C. (2026). Gene variants in periventricular nodular heterotopia. Acta Epileptologica, 8(1), Article 13. https://doi.org/10.1186/s42494-026-00243-9

Image Credits: AI Generated

DOI: 10.1186/s42494-026-00243-9

Keywords: periventricular nodular heterotopia, malformation of cortical development, epilepsy, FLNA, neuronal migration, genetic heterogeneity, radial glial cells, cerebral organoids, ARFGEF2, seizures, brain development, genetics