Researchers at the Baylor College of Medicine and the Jan and Dan Duncan Neurological Research Institute at the Texas Children’s Hospital have uncovered a new cell type in the human brain. The team used techniques including simultaneous electrophysiological and genomic profiling of single cells, and a new computational tool to discover that potentially a third of cells in human glioma fire electrical impulses. The studies showed that these impulses—action potentials (APs)—originate from tumor cells that are part neuron and part glia, supporting the concept that neurons are not the only cells that can generate electric signals in the brain.
The findings in addition indicated that patients with high relative numbers of the spiking hybrid cells (HCs) demonstrated improved survival outcome, hinting that the ratio of these cells may have potential prognostic value. The scientists also discovered that cells with hybrid neuron-glia characteristics are present in the non-tumor human brain, which they say highlights the importance of further studying the role of these newly identified cells in both glioma and normal brain function.
The research team, headed by Benjamin Deneen, PhD, and Rachel N. Curry, PhD, reported on their studies in Cancer Cell, in a paper titled “Integrated electrophysiological and genomic profiles of single cells reveal spiking tumor cells in human glioma,” in which they concluded “The implication of AP-firing non neuronal cells stands as a biological iconoclast, insofar as the prevailing tenets of neuroscience hold that neurons are the only cells capable of firing APs.” Daneen is Professor and Dr. Russell J. and Marian K. Blattner Chair in the department of neurosurgery, director of the Center for Cancer Neuroscience, a member of the Dan L Duncan Comprehensive Cancer Center at Baylor and a principal investigator at the Jan and Dan Duncan Neurological Research Institute.
“Gliomas are the most common central nervous system tumors with an estimated 20,000 cases diagnosed each year,” the authors wrote. Deneen added, “These tumors are universally lethal and have devastating effects on neurological and cognitive functions. Previous studies have shown that patient survival outcomes are associated with tumor proliferation and invasiveness, which are influenced by tumor intrinsic and extrinsic factors, including communication between tumor cells and neurons that reside in the brain.”
Researchers have previously described that glioma and surrounding healthy neurons connect with each other and that neurons communicate with tumors in ways that drive tumor growth and invasiveness. “… communication between tumor cells and their microenvironmental constituents has proven to be a critical mediator of glioma progression and is largely conducted by immunological and neural cellular components,” the scientists stated. However, they continued, “While tumor-neuron interactions have garnered significant attention, the electrophysiological profiles of tumor cells as they exist in situ within the human brain remain poorly defined.”
“We have known for some time now that tumor cells and neurons interact directly,” said first author Curry, postdoctoral fellow in pediatrics-neuro oncology at Baylor, who was responsible for conceptualizing the project. “But one question that always lingered in my mind was, ‘Are cancer cells electrically active?’ To answer this question correctly, we required human samples directly from the operating room. This ensured the biology of the cells as they would exist in the brain was preserved as much as possible.”
To study the ability of glioma cells to spike electrical signals and identify the cells that produce the signals, the team used Patch-sequencing (Patch-seq), a combination of techniques that integrates whole-cell electrophysiological recordings to measure spiking signals, with single-cell RNA-sequencing (scRNA-seq) and analysis of the cellular structure to identify the type of cells.
The electrophysiology experiments were conducted by research associate and co-first author Qianqian Ma, PhD, in the lab of co-corresponding author associate professor of neuroscience Xiaolong Jiang, PhD. This innovative approach has not been used before to study human brain tumor cells. “We were truly surprised to find these tumor cells had a unique combination of morphological and electrophysiological properties,” Ma said. “We had never seen anything like this in the mammalian brain before.”
“We conducted all these analyses on single cells. We analyzed their individual electrophysiological activity. We extracted each cell’s content and sequenced the RNA to identify the genes that were active in the cell, which tells us what type of cell it is,” Deneen said. “We also stained each cell with dyes that would visualize its structural features.”
Integrating this vast amount of individual data required the researchers to develop a novel way to analyze it. A streamlined computational framework capable of annotating glioma cells had yet to be developed, the authors noted, “… and cell annotation algorithms remain ill-equipped to assign integrated genomic and transcriptional profiles to single cells on a cell-by-cell basis. “To better define the HCs identified in our Patch-seq studies, we sought to create a computational platform that could annotate each cell from Patch-seq individually,” they wrote. “Because of the low cell numbers obtained using Patch-seq and the rarity of human glioma samples for use in these experiments, our annotation tool needed to be capable of analyzing each cell without a dependency on clustering methodologies, which requires hundreds to thousands of cells for optimal analysis.”
The researchers developed a new computation tool that could characterize the genomic and transcriptomic features of the individually recorded cells. “To define the spiking cells and determine their identity, we developed a computational tool—Single Cell Rule Association Mining (SCRAM)—to annotate each cell individually,” said co-corresponding author Akdes Serin Harmanci, PhD, assistant professor of neurosurgery at Baylor.
The collective results of the team’s studies showed that a subset of human glioma cells fire single, short action potentials and are defined by an amalgamation of GABAergic neuron and oligodendrocyte precursor cell (OPC) transcriptomes, which they termed GABA-OPCs. Harmanci stated “… the comprehensive data analyses revealed that the spiking hybrid cells in glioma tumors had properties of both neurons and OPC cells. Interestingly, we found non-tumor cells that are neuron-glia hybrids, suggesting that this hybrid population not only plays a role in glioma growth but also contributes to healthy brain function.”
In their study the investigators noted that previously, “… scientists identifying a class of spiking OPCs in the healthy rat brain posited that neurons are not the only cells capable of firing APs and suggested that an analogous population of spiking OPCs exists in human” Deneen et al believe that these previously described cell types are “… electrophysiologically equivalent to our HCs, which are transcriptionally defined by GABA-OPC signatures and represent heterogeneous group malignant and non-malignant cells.”
Curry pointed out, “Finding that so many glioma cells are electrically active was a surprise because it goes against a strongly held concept in neuroscience that states that, of all the different types of cells in the brain, neurons are the only ones that fire electric impulses. Others have proposed that some glia cells known as oligodendrocyte precursor cells (OPCs) may fire electrical impulses in the rodent brain, but confirming this in humans had proven a difficult task. Our findings show that human cells other than neurons can fire electrical impulses. Since there is an estimated 100 million of these OPCs in the adult brain, the electrical contributions of these cells should be further studied.”
Co-corresponding author Ganesh Rao, MD, Marc J. Shapiro Professor and chair of neurosurgery at Baylor, indicated that the results may have clinical importance. “The findings also suggest that the proportion of spiking hybrid cells in glioma may have a prognostic value,” Rao said. “The data shows that the more of these spiking hybrid glioma cells a patient has, the better the survival outcome. This information is of great value to patients and their doctors.”
The authors further concluded, “Whether GABA-OPCs with HC electrophysiology are endemic to the healthy human brain remains to be determined; however, given that OPCs are estimated to represent 3–4% of all gray matter cells and 8–9% of white matter cells in the mammalian brain, the cumulative neurophysiological contributions of these cells are poised to be significant and should not be ignored in either tumor or non-tumor contexts.”
Deneen noted, “The results offer an enhanced understanding of glioma tumors and normal brain function, a sophisticated bioinformatics pipeline to analyze complex cellular populations and potential prognostic implications for patients with this devastating disease … This work is the result of extensive equal collaboration across multiple disciplines— neurosurgery, bioinformatics, neuroscience and cancer modeling—disciplines strongly supported by state-of-the-art groups at Baylor.”
