Genetic Insights Into Hypothalamic Hamartoma: Unraveling Somatic Variants

Abstract

Objectives: Hypothalamic hamartomas (HHs) are rare developmental brain lesions associated with drug-resistant epilepsy and often subjected to epilepsy surgery. Brain somatic variants in genes affecting the Sonic hedgehog (Shh) and primary cilia signaling pathways have been implicated in approximately 50% of nonsyndromic HH cases. This study aims to characterize a new cohort of 9 HH cases and elucidate their genetic etiology.

Methods: We recruited 9 HH cases including 8 nonsyndromic cases of which 4 were type IV HH. Genomic DNA was extracted from peripheral blood and surgical brain tissues, and somatic variants were investigated using high-depth whole-exome sequencing.

Results: Pathogenic somatic variants in known HH genes (GLI3, OFD1, and PRKACA) were identified in 7 of the 9 cases. In addition, a 2-hit mutational event comprising a germline variant (predicted to impair kinase activity) and a somatic loss-of-heterozygosity was identified in TNK2, a gene encoding a brain-expressed tyrosine kinase.

Discussion: Our findings reinforce the role of somatic variants in Shh and cilia genes in HH cases while also shedding light on TNK2 as a potential novel disease-causing gene. This study emphasizes the increasing importance of brain mosaicism in epilepsy disorders and underscores the critical role of genetic diagnosis derived from resected brain tissue.

Figure 1. Representative MRI of HH Subtypes

(A) MRI T1 coronal sequence of patient ICM_239 with HH type II attached to one side of the hypothalamus (white arrowhead). (B) MRI T1 coronal sequence of patient ICM_242 with HH type III attached bilaterally to the hypothalamus (white arrowheads). (C) MRI T2 coronal sequence of patient ICM_212 with a giant HH type IV with bilateral attachment to the hypothalamus (white arrowheads) and ventricular enlargement.

Figure 2. Two-Hit Genetic Event in TNK2 and Protein Modeling of the p.M171T Variant

(A) Whole-exome sequencing (WES) and bioinformatic analysis workflow. (B) Top: somatic loss-of-heterozygosity (LOH) on chromosome 3q (chr3q29) detected in HH brain tissue (but not in the blood) from patient ICM_212 (B-allele frequency distribution). The absence of chromosome gain or loss (LogR ratio) on chr3q indicates copy-neutral LOH (cnLOH). Bottom: schematic of wild-type (WT) and mutant TNK2 loci in blood and HH brain cells. (C) In silico 3D modeling of the inactive (grey) and active (green) states of the TNK2 kinase domain. The p.M171 residue lies within a C-helix, whose inward switch is required for the activation of kinase domains. In the WT configuration, the M171 points toward the M201, displaying hydrophobic interactions (pink halo) that are lost in the mutant TNK2. The p.M171T is predicted to destabilize the inactive state of TNK2. ACMG = American College of Medical Genetics and Genomics; ADP = adenosine diphosphate; BAF, B-allele frequency; CNVs = copy number variants; HH = hypothalamic hamartoma; indels = insertions/deletions; LOH = loss-of-heterozygosity; LP = likely pathogenic; P = pathogenic; Ref, reference; SNV = single-nucleotide variant; VAF = variant allele frequency; VEP = variant effect predictor (Ensembl); VUS = variant of unknown significance.