Integration of Electrophysiological and Genomic Analysis Reveals Tumor Cells Capable of Generating Action Potentials in Human Gliomas

Background and Research Objectives

Gliomas are the most common malignant tumors in the central nervous system, with approximately 20,000 new cases annually. These tumors include two subtypes: isocitrate dehydrogenase (IDH) mutant (IDHmut) and IDH wild-type (IDHwt). IDHwt gliomas often have a poor prognosis, with a median survival of less than 14 months; in contrast, patients with IDHmut gliomas have a longer survival period. In recent years, advances in cancer neuroscience have revealed the crucial role of interactions between tumor cells and neurons in tumor progression, but the specific manifestations of tumor cells’ electrophysiological properties in tumor tissues remain unclear.

This study, led by scientists such as Rachel N. Curry from Baylor College of Medicine and other research institutions, integrates electrophysiology and single-cell genomics using “patch-seq” technology to perform an in-depth analysis of human glioma samples. The research aims to identify whether tumor cells possess action potential (AP) characteristics similar to neurons, analyze their gene expression profiles, and explore whether this property exists in both tumor and non-tumor tissues.

Source of Research

This study was published in the October 14, 2024, issue of the journal “Cancer Cell,” under the title “Integrated Electrophysiological and Genomic Profiles of Single Cells Reveal Spiking Tumor Cells in Human Glioma.” The paper’s main authors include Rachel N. Curry and Qianqian Ma, with the research team affiliated with Baylor College of Medicine and Texas Children’s Hospital, among other research institutions. The team hopes to provide new perspectives for the diagnosis and treatment of gliomas by uncovering the electrophysiological activity of tumor cells in vivo.

Research Process

Experimental Design and Procedure

1. Sample Collection and Preliminary Classification
   The research team obtained samples from brain glioma slices of ten patients, including seven IDH mutant gliomas, two IDH wild-type gliomas, and one non-tumor sample. Whole-cell patch-clamp recordings were performed on the slices, followed by single-cell RNA sequencing (scRNA-seq). Electrophysiological data were recorded from 157 cells, with 95 cells used for high-quality RNA sequencing analysis.

2. Electrophysiological Characteristics and Cell Classification
   Approximately 108 cells in the samples exhibited electrophysiological and morphological characteristics consistent with known neuronal cell types and were classified as pyramidal cells (PC), inhibitory neurons (IN), or non-spiking cells (NS). The remaining 49 cells exhibited neuronal electrophysiological properties, but their morphological characteristics were inconsistent with mature neurons. These cells were defined as “Hybrid Cells” (HCS) by the research team, constituting 36% in IDHmut samples and 30% in non-tumor samples, and were capable of generating single, small amplitude action potentials (AP).

3. Genomic and Transcriptomic Analysis
   To identify the genomic characteristics of these hybrid cells, the team conducted principal component analysis (PCA), revealing that these cells shared transcriptional features with both inhibitory neurons and oligodendrocyte precursor cells (OPC). For more precise cell type identification, the research team developed the Single Cell Rule Association Mining (SCRAM) tool. It annotates individual cells independently using a neural network model (NNM) and, when combined with single nucleotide variant (SNV) and copy number variation (CNV) analysis, generates genomic and transcriptomic profiles for each cell.

Results Analysis

1. Electrophysiological Characteristics of Hybrid Cells
   In IDHmut glioma samples, hybrid cells demonstrated high input resistance and were capable of generating small amplitude action potentials. PCA analysis showed that their electrophysiological properties were more similar to non-spiking cells (NSS).

2. Gene Characteristics and Tumor Markers
   Individual annotations of each cell using the SCRAM tool revealed that hybrid cells exhibited the transcriptomic characteristics of GABAergic neurons and OPCs, displaying the classical markers of IDH mutations. This study first revealed the widespread presence of these GABAergic OPCs (GABA-OPC) in IDHmut gliomas, comprising a significant proportion of tumor cells.

3. Relation between GABA-OPC and Patient Survival Rate
   The study analyzed the distribution of tumor cells with GABA-OPC characteristics in different IDH subtype tumors and found a higher proportion in IDH mutant gliomas. Kaplan-Meier survival analysis indicated that IDH mutant patients with high GABA-OPC scores had a significantly higher survival rate than those with low scores, suggesting that GABA-OPC characteristics might be associated with better prognosis in IDH mutant gliomas.

4. Potential Biological Mechanisms of GABA-OPC
   Genes related to voltage-gated sodium (Nav) and potassium (Kv) channels were detected in the transcriptome of GABA-OPC. Immunostaining further validated the expression of the sodium channel Nav1.1 in IDH mutant tumor cells. Electrophysiological validation using tetraethylammonium showed that Nav channels are a necessary condition for the action potential generation of GABA-OPC tumor cells.

Significance and Value of the Study

1. Revealing Electrophysiological Characteristics of Glioma Cells
   This study is the first to demonstrate in vivo that glioma cells have the ability to generate action potentials, a finding that challenges the traditional notion that only neurons can generate action potentials, paving the way for new directions in glioma cell research.

2. Association Between IDH Mutant Gliomas Prognosis and GABA-OPC
   The significant association between high GABA-OPC scores and improved survival rates in IDH mutant glioma patients suggests that the GABA-OPC characteristic may have a protective role in the IDH mutant subtype. This finding provides new markers for molecular typing and prognosis prediction in IDH mutant gliomas.

3. New Tool for Annotating Electrophysiological Characteristics
   The development of the SCRAM tool offers an accurate method for single-cell annotation, useful for identifying and analyzing specific genomic and transcriptomic features of tumor and non-tumor cells, particularly effective in processing small sample sizes. This tool helps advance the application of single-cell RNA sequencing in complex tumor tissues.

Conclusions and Outlook

The findings of this study provide new perspectives on the complex biological characteristics of glioma cells. The action potential generation of GABA-OPC tumor cells may have a significant impact on the tumor microenvironment. Future research could further explore whether tumor cells form functional connections with surrounding neuronal networks and whether this interaction affects the risk of seizures in patients. Additionally, the existence of action potential-generating GABA-OPCs in non-tumor brain tissue, as discovered in this study, suggests that these cells may have previously undiscovered roles in normal brain function.