It is increasingly clear that the nonneuronal brain cells called glia are intricately involved in the neuronal crosstalk at synapses.

In terms of brain function, the synapse is where it all happens. It is at these molecular junctions— 100 trillion or more of them, by current estimates—that neurons talk to one another, exchanging biochemical messages that ultimately orchestrate how we feel, think, remember, and behave. Now, slowly and cautiously, neuroscientists are rewriting the science of the synapse. The classic, textbook view of nerve transmission at the synapse depicts an electrical signal traveling along a nerve fiber and triggering the neuron to release chemical transmitters, which then diffuse across a tiny cleft and lock onto receptors on a neighboring neuron. The nerve signal is thus transmitted from one neuron to the next, like the baton in a relay race. But it is turning out that this tidy, two-party picture may not be telling the whole story. It is increasingly clear that the nonneuronal brain cells called glia are intricately involved in the neuronal crosstalk at synapses. As scientists probe deeper, armed with new experimental techniques, powerful imaging tools, and a better understanding of glia than ever before, a new view of the synapse is emerging. Now Starring: the Astrocyte Central to this new view is the concept of a “tripartite synapse,” in which the astrocyte, a specialized type of glia, is an integral part of the triad. Some scientists are convinced that glia, long shunned as the mere “glue” of the nervous system, as their Greek name implies, in fact play critical, active roles in modulating neuronal transmission. Preliminary data suggest that glia may be important regulators of synaptic plasticity, the process of synapse strengthening or weakening believed to underlie learning and memory. Growing evidence suggests a new starring role for cells that were previously viewed as little more than a supporting cast to neurons. Phil Haydon, a molecular neuroscientist at the University of Pennsylvania, says researchers are becoming aware that there are physiological functions in which astrocytes do more than provide support. “They are not just listening to neurons, but they talk and instruct the neurons as well,” Haydon says. Haydon published new results in the Oct. 7 issue of Science that indicate astrocytes are intricately linked in the regulation of synaptic strength and plasticity. Moreover, the work identifies a specific molecular pathway through which astrocytes exert at least some of their effects at the synapse. Using a mouse model genetically altered to selectively block transmitter release from astrocytes, Haydon’s group found direct evidence that astrocytes regulate synaptic transmission and modulate plasticity at the synapse by controlling levels of adenosine in the space surrounding synaptic junctions. An important neurotransmitter implicated in the sleep-wake cycle and in energy regulation in the cell, adenosine is a byproduct of adenosine triphosphate (ATP), which is released by astrocytes in response to neural signaling. “These studies place the astrocyte at center stage in the control of adenosine,” the authors wrote. As such, the work provides an important new chapter in scientific understanding of the cellular machinery by which astrocytes control synaptic communication. “My view at the moment is that the astrocytes integrate information from the synapse and then can provide feedback to modulate those synapses,” Haydon says, which ultimately influences the overall strength of the synaptic signal by either strengthening or weakening it. Stanford neurobiologist Ben Barres agrees that, while there is still much to learn, “There is an emerging feeling, based on a building body of literature, that glia are participating actively in synapses. We’re starting to see similar effects [of glia on synapses] pretty much everywhere people look.” Precisely how glia are participating in synapses is an open question, but research published by Barres’ laboratory in February in Cell provides a clue. Thrombospondin (TSP), a protein his team identified that is secreted by astrocytes, may represent one mechanism by which glia influence synapses. “Thrombospondin is the first known protein that is sufficient to trigger synapse formation when added to neuron culture dishes,” Barres says. The protein is expressed in the brain during a specific window of nervous system development, during which it seems to act as a “permissive switch” to induce the formation and development of synapses. The TSP gene is one of only a few that are far more active in humans than in our primate cousins, according to emerging data from human and monkey genome comparisons. This suggests that the TSP protein may be a key to humans’ ability to form more synapses—and, by extension, a clue to our capacity for higher cognition. A Role in Disease? If glia are so intimately involved in synaptic transmission and plasticity in the normal brain, might they also have roles in disease? Increasing evidence suggests that glial cells—not just astrocytes but their cousins oligodendrocytes and microglia as well— contribute significantly to pathological processes in some neurological diseases, including multiple sclerosis, epilepsy, chronic pain, spinal cord injury, and neurodegenerative disorders such as Alzheimer’s and amyotrophic lateral sclerosis (Lou Gehrig’s disease). Moreover, through their control of critical neurotransmitters such as glutamate, glial cells are also being scrutinized for possible involvement in psychiatric disorders, including anxiety, depression and schizophrenia. “We’re starting to realize we need to pay attention to what glia are doing not only normally but in disease states,” says Barres. “I’d be very surprised if there’s any disease in the brain in which glia don’t play an important role.” Proving an Assumption To Shai Shaham, a Rockefeller University neurobiologist who studies glia in the worm C. elegans, the evidence is now strong for what he says has been a central assumption among glia researchers: “Glia must function, at least in part, to regulate neuronal parameters.” Writing in the September Neural Development issue of Current Topics in Developmental Biology, Shaham says the assumption made sense, given that glia are ubiquitous in the nervous system and that they entwine and ensheathe nerve fibers in the brain and periphery. But doing the scientific experiments that would answer the question of just how glia are affecting neuronal transmission—and vice versa—was inordinately difficult because without glia, neurons die. “If you want to study the role of glia in the brain, you’d want to get rid of the glia and see if behavior changes,” Shaham explains. “The problem is, if you get rid of the glia, you will also get rid of the neuron. That, in my mind, has been the major stumbling block in trying to understand what other functions glial cells have, besides neuron-survival functions.” Researchers have employed a variety of innovative approaches to get around this central problem. Barres’ team, for example, has developed an experimental system in which neurons (retinal ganglion neurons from the rat, in this case) are kept alive without glia by growing the neurons in culture dishes conditioned with growth factors. Researchers can then add glial cells back into the cultures gradually and observe how neuronal activity changes. Haydon’s group and others have developed a method called photolysis to selectively “tickle” individual astrocytes into signaling while recording associated changes in the neuron. Or, as Haydon puts it, “to make the astrocyte start talking, and then see what the consequences are for the neuron.” In C. elegans, Shaham’s team has found that glia are not required for neuron survival. “We can use lasers to selectively ablate individual glia and ask what happens,” Shaham says. So far, his team has found various defects in the absence of glia, including behavioral changes indicating a disruption in neuron signaling, abnormal positioning of neurons, and, most dramatically, structural changes in the dendritic tips of neurons. “If you get rid of the glia, the whole elaborate structure that the tip normally takes on just disappears,” Shaham says. A Decade of Awakening The current flurry of scientific interest in astrocytes, and the reason their functions have been so mysterious for so long, is linked to the relatively recent ability to measure the activity of these cells at all. Methods to record electrical currents from neurons have been available for more than a century, but astrocytes are fueled by calcium, not electricity. It was not until the early 1990s that scientists could monitor the signals astrocytes were sending, thanks to new confocal microscopy and imaging approaches that could record changes in calcium levels. The improved techniques proved fruitful in short order: in 1994, four separate research groups published evidence that astrocytes can signal neurons directly. Haydon’s group showed that, in response to calcium influx, laboratory-grown astrocytes and Schwann cells (glial cells found in the peripheral nervous system) release glutamate, the most prevalent excitatory neurotransmitter in the brain. It was known that astrocytes contributed to the regulation of glutamate levels at the synapse by secreting molecules that whisk away excess glutamate and that glial uptake of neurotransmitters is essential for proper synaptic activity, but direct release of glutamate was something only neurons were supposed to do. “It was a real awakening,” says Haydon, and it stimulated a wave of additional research to figure out what these cells were doing. “In the past decade, exquisite, beautiful work has gone on to really test the idea that astrocytes listen to neurons and that they talk back. Now, I think there’s overwhelming evidence that this occurs.” It is now clear that astrocytes respond to neural activity by releasing internal stores of calcium, and this calcium signal unleashes neurotransmitters that signal and modulate the synapse. This biochemical feedback to the neuron can, in turn, regulate how much transmitter the neuron itself releases or alter the firing properties of the neuron. As such, the astrocyte functions as a kind of neuronal chaperone, monitoring the interactions between neurons and releasing signals that can regulate these interactions. Permitting Kissing—Or Not “You have these two neurons coming together at the synapse, and whether or not they keep kissing one another depends upon whether the astrocyte is telling them it’s okay,” says Mark Ellisman, a molecular neurobiologist at the University of California, San Diego. “There are now a whole bunch of open questions about how this interplay works when it’s working right, and how it goes wrong in disease conditions.” With so many questions still unresolved, experts say it is too soon to rewrite the textbooks on synaptic transmission, but cautious optimism prevails. “I think the study of the astrocyte’s role in the synapse is sort of where the study of the neuron was in the 1930s,” Haydon says. “We’ve been working on it for a decade and now there are so many questions, and many people are excited by these questions. My anticipation is that in the next five years, we will really be able to understand what this listening and talking by the astrocyte means.”

synapse, glia, neuron, synaptic, plasticity, transmission, chemical, transmitter, neurotransmitter