One of the most well-studied synapses in the brain continues to surprise neuroscientists. According to a May 18 study in Nature Communications, mossy fiber synapses, so named because their terminals look a bit like moss growing on the axons, have an unexpected way of regulating the flow of information in the hippocampus: the postsynaptic cells that receive neurotransmitter signals can release their own glutamate to tamp down the transmission from the cell on the presynaptic side.
This so-called retrograde signaling was totally unexpected and depends on calcium influx to the postsynaptic cell, meaning researchers might have to rethink the results of past experiments that used in vitro conditions with different calcium availability.
The findings are “a big deal” for neuroscientists, says Chris McBain, a synaptic physiologist at the National Institutes of Health who was not involved in the study. “Retrograde glutamatergic signaling is a really rare occurrence in the central nervous system,” he notes, and to find it in mossy fibers “adds another layer of complexity onto one of the most complex synapses.”
Mossy fiber synapses: Big players in the hippocampus
The researchers behind the new paper, led by neurophysiologist Peter Jonas of the Institute of Science and Technology Austria, were investigating the plasticity of hippocampal neurons, the dynamic changes in connections between cells that contribute to the functioning of neural circuits and that ultimately underlie learning, memory, and other cognitive abilities. János Szabadics, a neurophysiologist at the Institute of Experimental Medicine, Budapest, puts it quite simply: “Without synaptic plasticity, the brain would be just a bag of wires,” he says.
Jonas’s team was particularly focused on the physiology of post-tetanic potentiation (PTP), a phenomenon when, for seconds to minutes after a high-frequency burst of stimulation, the amount of neurotransmitter released into the synapse is increased. PTP is one of the main mechanisms of plasticity in mossy fiber synapses, which play a key role in relaying information in the hippocampus.
Mossy fiber synapses have also been studied as a model for synapses in general because the bulbous tip, or terminal, of the cell’s axon that releases vesicles of neurotransmitters is several microns in diameter, and thus easier to find and manipulate than those of most neurons. In fact, McBain says, mossy fiber synapses are so well-studied that “you’d think that all the surprises have been found, and that what’s left is sort of nuanced . . . . So, for [this research team] to come along and show so cleanly that a change in the postsynaptic calcium actually can trigger a presence or absence of this post-tetanic potentiation I think is remarkable.”
It wouldn’t surprise me at all if there are different manifestations or flavors of this phenomenon that are all over the place.—Chris McBain, National Institutes of Health
According to co–first author Yuji Okamoto, the discovery was somewhat serendipitous. Both Okamoto and David Vandael, the study’s other first author, are postdocs in Jonas’s lab and were using a technique Jonas pioneered called “paired recording,” where electrophysiological measurements are concurrently taken of the postsynaptic cell and presynaptic terminal on either side of a synapse.
Last year, Jonas’s team demonstrated that PTP stems from an increase in the presynaptic terminal’s readily releasable vesicle pool—the amount of neurotransmitter packaged and ready to go when the neuron is activated. In their latest work, they were keen to determine the rules governing the formation of this neurotransmitter cache.
One question they had was whether PTP induction is facilitated by activity in the postsynaptic cell, such as would come from it being stimulated by its many other connections with different cells. Known as associative synapse plasticity, this pattern of regulation often relies on an increase in calcium in the postsynaptic cell.
To test for this, the team plunged an electrode into the postsynaptic cell’s interior and reduced the amount of calcium-binding compounds in the buffer solution, thereby making more calcium available to the cell. If anything, they expected this boost in available calcium would increase the degree of PTP seen after the induction stimulus—100 quick jolts to the presynaptic terminal and the postsynaptic cell being observed. Instead, and quite unexpectedly, the potentiation was reduced or completely disappeared.
To zero in on calcium’s role, the researchers did the opposite experiment and essentially drained calcium from the postsynaptic side by dialing up the calcium-binding compounds—and PTP returned. That means “the calcium influx into the postsynaptic cell somehow blocks the potentiation,” says Okamoto. In other words, when calcium is available, the postsynaptic cell can prevent PTP, and therefore meter how much information, in the form of neurotransmitter, it receives from a given synapse.
“We were, of course, surprised,” he says. “It’s really weird.”
The surprises continued as Okamoto and his colleagues dug deeper. First, they tested whether the postsynaptic cell was stopping PTP by signaling the presynaptic terminal with endocannabinoids, as retrograde endocannabinoid signaling occurs in other synapses and is the most well-established form of retrograde synaptic signaling. But bathing the presynaptic terminals in a compound that blocks cannabinoid receptors had no effect.
Instead, the team tried treating the presynaptic terminal with blockers for the glutamate receptors known to be highly abundant on mossy fiber axons. That was enough to recover PTP knocked out by increased calcium availability on the postsynaptic side. This suggests that under conditions with abundant calcium, glutamate from the postsynaptic cell can travel back to the presynaptic terminal to inhibit PTP.
This kind of retrograde signaling wasn’t thought to happen in mature mossy fiber synapses at all, Okamoto says, especially not with glutamate.
Szabadics describes the methodology as “very clever,” adding that “there are not too many laboratories in the world” that can perform the kind of detailed electrophysiology experiments in the new paper.
A glutamate mystery solved
In some ways, the findings actually help answer long-standing questions in synapse biology, says McBain. The presence of glutamate receptors at presynaptic terminals was established decades ago, but no one was certain where the glutamate that stimulates them comes from. Many presumed that some of the glutamate released by the presynaptic terminal into the synapse diffuses back, particularly if a lot of the neurotransmitter is released, he says. And while that could still happen at times, “now you’ve got a mechanism that sort of flies in the face of that,” he says, “and it sort of makes sense that you have this really nice barometer of postsynaptic activity” in the form of postsynaptic glutamate release. “It wouldn’t surprise me at all,” he adds, “if there are different manifestations or flavors of this phenomenon that are all over the place.”
At this point, researchers can only guess as to why this kind of feedback mechanism evolved, says Szabadics. One idea is that it could help ensure important memories aren’t overwritten by less useful data; but that or any other potential explanation at this point is “very, very speculative,” he cautions.
The fact that PTP is regulated by the calcium available to the postsynaptic cell has far-reaching implications for interpreting past neurophysiological research, McBain notes. “One of the inconvenient truths of this paper is that if you go back and look at a lot of other studies that have been done at this synapse,” you’ll find that the buffers used for electrophysiological recordings vary, meaning calcium availability has too, he says. This might actually explain why labs often observe different degrees of PTP from one another, he adds.
This lack of methodological standardization arose because “nobody was thinking that postsynaptic calcium is going to be a confound in establishing the basic parameters of the synapse,” McBain says. “Everybody needs to sort of go back and rethink what is the calcium dependence of most of the basic parameters that we knew about the synapse.”