Article In Brief
A leading investigator outlined advances in understanding gliomagenesis and prospects for new treatments, including a metalloproteinase that inhibits the growth of gliomas.
Myelination is a central process of early brain development, and work in the past decade has drawn a tight link between dysregulation of myelination and development of gliomas of childhood. Much of that work was led by Michelle Monje, MD, PhD, who has now begun to link normal myelination to neuronal activity, and to show that growth of gliomas is accelerated by the activity of the neurons they invade.
Dr. Monje outlined advances in understanding gliomagenesis and prospects for new treatments, based on these advances, in the Sidney Carter Award in Child Neurology lecture at the AAN Annual Meeting in May.
To understand gliomas of childhood, it is necessary to better understand the normal drivers of myelination in the developing brain, said Dr. Monje, associate professor of neurology at Stanford University. “We hope to understand the way neuronal activity influences glial cells, and the way that interaction is dysregulated and hijacked in the context of glioma.”
Myelination is the job of oligodendrocytes, which arise from oligodendrocyte precursor cells (OPCs). “Many pediatric gliomas resemble OPCs on the molecular level,” she said, as shown by transcriptional analysis, pointing to a 2018 paper in Science.
“The spatiotemporal pattern of gliomagenesis in the childhood, adolescent, and young adult brain maps fairly well onto discrete stages of development, such that at a time when there is a discrete wave of myelination in mid-childhood,” she said, that there is also a spike in the development of diffuse intrinsic pontine glioma.
“Active myelination in the cerebral lobes, particularly in the frontotemporoparietal lobe, coincides with the highest risk for pediatric high-grade glioblastoma,” she said.
Work in the 1990s by the late Benjamin Barres, PhD, introduced the idea that “neurons themselves may regulate the extent to which they become myelinated,” Dr. Monje said, an idea that “has been highly controversial in the myelin field for many years.”
Optogenetics As a Tool
To explore this possibility, Dr. Monje turned to optogenetics. In this technique, a light-sensitive cation channel is expressed in cortical projection neurons in mice, which can then be stimulated by light delivered through an implanted optical fiber.
In response to stimulation and the elevation of complex motor activity that it induced, “we found that there was a rapid and robust increase in the proliferation of OPCs in the target circuit,” whose number more than quadrupled in the target area, resulting over time in an increase in the number of newly generated oligodendrocytes and a thickening of the myelin sheath—a finding reported in 2014 in Science.
“The activity-regulated changes in myelin structure would be predicted to influence impulse conduction and overall function, and we observed that mice that had undergone this plasticity-dependent myelination had exhibited improved motor performance,” she said. Recently, she has found that the myelination effect depends on BDNF-TrkB signaling, and that loss of TrKB—a receptor for brain-derived neurotrophic factor (BDNF)—impaired learning. Most recently, her lab has shown in a report in Neuron in May that a reduction in adaptive myelination contributes to the cognitive effects of chemotherapy.
“There is the potential to leverage these insights to promote new myelination,” as a treatment in adult demyelinating diseases, she said.
Given the importance of neuronal activity to adult myelination, and the close connection between normal and disease myelination processes, Dr. Monje wanted to know whether activity also promoted pediatric glioma proliferation. “This is suggested by the histopathological hallmark of these high-grade gliomas,” called perineuronal satellitosis, she said, in which high-grade gliomas tend to form a tight association with neurons.
In 2015, her lab reported in Cell that when human glioma cells were transplanted to the premotor area and treated with optogenetics, they increased their proliferation and growth. “So brain activity can promote the growth of these cancers,” she said.
The mechanism involved the release of secreted factors, she found, and the same factors promoted multiple types of glioma, including pediatric glioblastoma, diffuse intrinsic pontine glioma, adult glioblastoma, and anaplastic oligodendroglioma.
Dr. Monje identified BDNF as one important mechanism mediating neuron-glioma interactions, as well as another candidate protein, which unexpectedly turned out to be a well-known presynaptic adhesion molecule called neuroligin-3. She showed it was cleaved at the membrane by a metalloproteinase called A Disintegrin and metalloproteinase domain-containing protein 10, or ADAM10, and released, a process promoted by neuronal firing. Neuroligin-3 not only promoted glioma, but was unexpectedly required for glioma progression.
“Inhibiting ADAM10 blocks glioma growth,” Dr. Monje said. “This makes it a really intriguing therapeutic target. There is an ADAM10 inhibitor that is brain penetrant, and its use blocks tumor growth across multiple types of glioma. We are going to be bringing this inhibitor forward for a phase 1 clinical trial, hopefully by late 2019.”
Why is this synaptic protein so important for regulating glioma growth? Dr. Monje has shown that after activity-regulated increase in cleavage and secretion of neuroligin-3, “there is a robust stimulation of numerous oncogenic pathways in the glioma cell.”
Among the pathways upregulated are those for synapse-related genes, including glutamate receptors and synaptic structural molecules. “Just like there are axon-glia synapses between neurons and OPCs, could there similarly be axon-glioma synapses?” It is a “heretical idea,” and one that Dr. Monje's lab is exploring.
“The mechanistic parallels evident in normal and malignant neuron-glial interactions underscore the extent to which these cancers are hijacking normal mechanisms of neural development and neuroplasticity, and highlight that we need to approach this tumor type from the perspective of neuroscience,” Dr. Monje said. “Hopefully, therapeutic targeting of these crucial mechanisms will achieve improved outcomes for our patients.”
“It remains to be seen how safe it is to interrupt this pathway, given that it is involved in normal myelination and remyelination processes,” commented Lisa DeAngelis, MD, FAAN, chair of neurology and acting physician-in-chief at Memorial Sloan-Kettering Cancer Center in New York. There may be some latitude because blocking neuroligin-3's effect would presumably impair, but not kill, progenitor cells. “But we will have to determine that,” and the safety may differ between adults, for whom myelination is largely complete, and children, she said.
“What is really exciting about this work is that this neuroligin-3 interaction between tumor cells and neurons appears to be agnostic between tumor types,” she said. “This means that potentially, any treatment that can interrupt this connectivity feeding the tumor could be useful across the spectrum. That is a very different approach than the current emphasis on molecular subtyping of tumors.”
Dr. Monje has received personal compensation for consulting, serving on a scientific advisory board, speaking, or other activities with Cygnal Therapeutics.