Combined patch-clamp electrophysiology and single-cell genomic analysis reveal spiking tumor cells at the neocortical glioblastoma interface in humans
Authors:
- Tong TongTTTong Tong
- Josephine D HendriksenJHJosephine D Hendriksen
- Kirstine J ElbækKEKirstine J Elbæk
- Kata MolnarKMKata Molnar
- Freja V ChristiansenFCFreja V Christiansen
- Attila OzsvarAOAttila Ozsvar
- Jens T EschenJEJens T Eschen
- Francisco G Rodríguez-GonzálezFRFrancisco G Rodríguez-González
- Ilayda D PusatIPIlayda D Pusat
- Emilie Littau ChristensenECEmilie Littau Christensen
- Mads RahbækMRMads Rahbæk
- Kathrine Pii FrederiksenKFKathrine Pii Frederiksen
Open Access
- 19 March 2026
- journal article
- Published by Oxford University Press (OUP) in Neuro-Oncology
- Vol. 28 (6) , 1442-1457
Abstract
Background: Electrophysiological features of glioblastoma cells (GBCs) remain largely elusive, challenging our comprehension of glioblastoma pathophysiology. Spiking GABAergic-oligodendrocyte-progenitor (OPC) tumor cells were recently described in IDH-mutant glioma, correlating with prolonged patient survival. Here, we characterize single-cell features at the neocortical leading edge (LE) of glioblastoma patients using combined electrophysiological, morphological, and transcriptomic profiling. Methods: We examined GBCs and non-tumor cells using acute and cultured organotypic slices of cancer-infiltrated neocortical tissues from glioblastoma patients. Electrophysiological properties of LE cells were investigated using whole-cell patch-clamp recording, with dye loading to characterize single-cell morphology. We used Patch-seq to determine the transcriptomic features of recorded LE cells and discriminate tumor and non-tumor cells, followed by gene set enrichment analysis and CellChat to identify differential gene expression and signaling. Results: Upon depolarization, more than half of LE cells show aberrant action potentials (aAPs), akin to neurodevelopmental cells. Reconstructed LE cells have abnormal somatodendritic morphology. Patch-seq revealed that GBCs and non-tumor cells share a similar electrophysiological phenotype, including aAP generation, depolarized membrane potential, and elevated input resistance. Transcriptomic analysis shows that the aAP phenotype occurs across diverse GBC states and correlates with lower enrichment of proliferation-related pathways at the single-cell level, but higher enrichment of inflammatory/immune, angiogenic, and mesenchymal transition pathways. Non-tumor cells exhibit hybrid transcriptomic signatures, with predominantly neuronal but minor enrichment of astrocytic features. Conclusion: We find electrophysiological aAP behavior of GBCs in human glioblastoma, closely resembling that of hybrid cells in IDH-mutant glioma, supporting the hypothesis of neuronal mimicry across different glioma types.Keywords
Funding Information
- Kræftens Bekæmpelse (R295-A16770)
- Kræftens Bekæmpelse (R304-A17698-B5570)
- Aarhus University Research Fund (AUFF-E-2023-9-41)
- Kræftfonden; the Independent Research Fund Denmark (4285-00323B)
- H. Lundbeck A/S
- National Institutes of Health (R325-2019-1490)
- Lundbeck Foundation (37741)
- Danmarks Frie Forskningsfond
- Krogh Invest and Danish Cancer Society (R295-A16770)
This publication has 52 references indexed in Scilit:
- Glioblastoma hijacks cholinergic networksCancer Cell, 2025
- Integrated electrophysiological and genomic profiles of single cells reveal spiking tumor cells in human gliomaCancer Cell, 2024
- Glioblastoma hijacks neuronal mechanisms for brain invasionCell, 2022
- Integrated analysis of multimodal single-cell dataCell, 2021
- Oligodendroglial GABAergic Signaling: More Than Inhibition!Neuroscience Bulletin, 2021
- An Integrative Model of Cellular States, Plasticity, and Genetics for GlioblastomaCell, 2019
- The Cerebral Cortex2018
- Neuronal Activity Promotes Glioma Growth through Neuroligin-3 SecretionCell, 2015
- GABAergic disinhibition and impaired KCC2 cotransporter activity underlie tumor-associated epilepsyGlia, 2014
- Integrated Genomic Analysis Identifies Clinically Relevant Subtypes of Glioblastoma Characterized by Abnormalities in PDGFRA, IDH1, EGFR, and NF1Cancer Cell, 2010