Synapses: The cells behind the scenes
Non-neural cells drive the formation of neural connections in the vestibular system
By Parizad Bilimoria
Gabriel Corfas (back left), Maria G?mez-Casati (back right) and Joshua Murtie (front) have found that non-neural cells are essential for forming neural connections—known as synapses—in a living nervous system.
Credit: Ethan Bickford
Glial cells are the Cinderellas of neuroscience. Although their beautiful variety and structural intricacy were noted over a century ago by Nobel laureate Ram?n y Cajal, these abundant nervous system cells were until recently cast into the shadows—thought of as mere servants for neurons, the electrically active components of neural circuits. Indeed, glia can nourish neurons, help them relay messages and mop up their spills.
But in the last decade, glia have arrived at the ball, owing largely to the discovery that they play a dynamic role in communication between neurons. A recent study by the laboratory of Gabriel Corfas, PhD Professor of Neurology at Children’s, demonstrates for the first time—in a living nervous system—that non-neural cells are essential for forming the neural connections known as synapses. It also uncovers some of the molecular crosstalk between these cells and the neurons they connect. Although the study used the vestibular system in the inner ear as a model, this improved understanding of synapse development may shed light on signaling pathways implicated in a number of brain disorders, especially psychiatric disorders.
Studying mice, Corfas and postdoctoral fellows Maria G?mez-Casati, PhD, and Joshua Murtie, PhD, focused on the utricle, a part of the inner ear that senses the body’s horizontal movements, aiding our sense of balance and orientation. Hair cells in the utricle detect the flow of tiny crystals in viscous fluid, and convey this information to the brain through synapses with sensory neurons.
But Corfas and his colleagues found that these synapses don’t form without instruction from “supporting cells” (analogous to glia elsewhere in the nervous system) that surround the hair cells and sensory neuron terminals. Moreover, the supporting cells in turn take their cue from the sensory neurons, showing that synapse formation requires a dynamic crosstalk between neurons and glia or glia-like cells.
No crosstalk, no connections
Top: Anatomy of the inner ear. Together, the vestibular organs—the utricle, saccule and semicircular canals—sense changes in the body’s movements and orientation and send this information to the brain via the vestibular nerve. Here, hair cells in the utricle, showed up close in the inset, sense changes in horizontal motion, by detecting the flow of tiny crystals in a viscous fluid. They send this information to the brain through connections, or synapses, with sensory neurons. Supporting cells, analogous to glia elsewhere in the nervous system, surround these hair cells and sensory neuron terminals.
Bottom: How supporting cells drive synapse formation: Sensory neurons produce the growth factor neuregulin (NRG), with activates ErbB receptors on supporting cells. This activation leads supporting cells to secrete their own signal—brain derived neurotrophic factor (BDNF). The BDNF drives synapse formation, and possibly synapse maintenance, by acting on the hair cells (and perhaps the sensory neurons too). Signals traveling from the sensory neuron to the hair cell may also contribute.
CREDIT: Graham Paterson (based on original image from Corfas Lab)
Corfas and others had already learned that sensory neurons produce the growth factor neuregulin (NRG), which activates so-called ErbB receptors on the supporting cells. Working with collaborator M. Charles Liberman, PhD, of the Massachusetts Eye and Ear Infirmary, they decided to see what would happen if this messaging were disrupted. When they blocked ErbB function in the glial cells of the peripheral nervous system (including vestibular supporting cells), results were dramatic.
“These mice have very significant difficulties in walking normally,” Corfas says. “They move around shaking their heads, a classic sign of vestibular defects, and they don’t really know what is up or down. In a swimming test, the animals have no idea where the surface is, and they dive, something that a normal mouse would not do.”
Off the balance beam
This video shows the vestibular defects of mice lacking proper ErbB function in glial cells of the peripheral nervous system (including the supporting cells of the utricle). Note the poor coordination, head shaking, circling behavior, trouble walking on the balance beam and difficulty swimming.
CREDIT: Video reproduced courtesy of PNAS. Original citation: Gómez-Casati et al. PNAS. 2010 Sep 28; 107(39):17005-10.
The animals’ brain and inner ear anatomy seemed normal, with no signs of missing or disorganized cells. And hair cells seemed to be functioning normally, converting mechanical stimulation into chemical activity.
But a closer look at the utricle revealed an almost complete lack of synapses. “It was mind-blowing,” Corfas recalls. “We had been trying to find a defect for months and months and could not find it. And suddenly it was so dramatic, finding that these very critical yet minute structures—the synapses—were the key.”
How did the ErbB defect in the supporting cells lead to this loss of synapses? Could the glia-like supporting cells be sending their own messages? Corfas’s team looked for changes in the factors secreted by these cells. Testing a panel of growth factors known to be important in the inner ear, they zeroed in on brain-derived neurotrophic factor (BDNF), and showed that BDNF production in supporting cells is diminished by the ErbB defect. When the researchers restored BDNF in vestibular supporting cells—through a genetic trick—both synapse formation and vestibular function were also restored.
Corfas observes that it is in some ways a “paradigm shift” to think of glial BDNF as a driver in synapse development. While BDNF made in neurons was long known to orchestrate synapse development, this was actually the first time BDNF from non-neural cells was implicated.
The ear as a window into the brain
Corfas’s group continues to study NRG-ErbB control of inner ear development. With Liberman, the team had found that supporting cells in the cochlea, a key inner ear structure involved in hearing, also use NRG-ErbB signaling to promote the survival of sensory neurons. Because loss of sensory neurons in the cochlea has been associated with hearing loss in humans, these findings could potentially be helpful to patients.
“What we are starting to learn about the inner ear—both the cochlea and the vestibular system—is going to open new doors to thinking about ways to help people with auditory and vestibular dysfunction,” Corfas says. “These are very common disorders, and they have a significant impact on quality of life.”
The new findings also elicit many questions about synapse development in the brain, and may serve as a stepping stone for understanding psychiatric disorders. Both NRG-ErbB and BDNF signaling have been linked genetically to neuropsychiatric disorders and may be involved in their development, Corfas says.
In 2007, another study by the Corfas lab, led by Kristine Roy, PhD, and Joshua Murtie, PhD, revealed that blocking ErbB signaling in oligodendrocytes, a group of glia of the central nervous system, caused mice to develop features suggestive of psychiatric disorders like schizophrenia.
In addition to abnormal social behavior, these mice displayed other features associated with schizophrenia: Their axons, the extensions of neurons that carry signals to other neurons, had less myelination (insulation), slowing information transfer. The mice also had abnormalities in neuron-to-neuron communication mediated by the chemical dopamine. (Existing drugs for schizophrenia act on the dopamine pathway.)
The findings in the vestibular system suggest new avenues for investigation of schizophrenia: Might synapse problems contribute to the abnormal behaviors of these mutant mice? Is missing glial BDNF responsible for any of the changes?
Clearly, it’s a new age for glia, which are turning out to be just as important as neurons in synapse formation, in addition to neurons’ very survival. Without neuron-glia crosstalk, we might be unable to walk, or have trouble hearing or even socializing. Stay tuned to the Corfas lab to find out what else glia do for us, and how they do it.