Dendritic Spines Don’t Go with the Flow

Spines’ Shape Shifting May Control Communication and Underlie Independence of Synapse

Neurons receive incoming signals through synapses at hundreds of dendritic spines, the lollipop-shaped structures with thin necks and bubblelike heads that stud the surface of dendrites. Each spine serves as an antenna relaying the chemical and electrical signals at the synapse to the cell body. If the din is loud enough, the entire cell will rouse itself to fire an action potential.

Photo by Graham Ramsay

Brenda Bloodgood and Bernardo Sabatini found that synapses may be regulated by their isolation: the narrow passage that connects them to the cell clamps down in response to certain stimuli.

Synapses hold the key to understanding how the brain perceives, records, and responds to incoming information. With the right stimulation, some of the synaptic signals grow stronger, like soloists in a chorus. And this regulation of synaptic strength allows the brain to change in response to experience.

Shape Up or Ship Out
Many studies have looked at the complex molecular changes that influence synaptic strength. But a study led by Bernardo Sabatini, HMS assistant professor of neurobiology, suggests that part of the control may lie in the shape of the spines themselves. He and graduate student Brenda Bloodgood found that the necks of dendritic spines constrict or widen in response to different inputs, regulating the ability of molecules to flow from the spine into the cell body. This action, detailed in the Nov. 4 Science, could be a way that the spines control synaptic strength and give synapses some independence from the cell.

“One of the big questions in neuroscience is, how do neurons integrate all the synaptic inputs they get?” said Bloodgood. “Not all synapses on a neuron are equal.” The structure of dendritic spines keeps each synapse separate, marooned on its own peninsula at the cell surface. It is thought that this physical separation helps regulate the synapses, allowing each one to keep its own pool of molecular signals. But until now, it was difficult to study whether their isolation was a regulated property.

Diffusion Gets Green Light
Sabatini’s lab has developed techniques for watching the events of a single synapse using a dual-laser two-photon scanning microscope (see Focus Nov. 12, 2004). In this study, the researchers transfected rat neurons of the hippocampus with a red fluorescent protein and a green one capable of being photo-activated. The first laser lit up the red throughout the cell, illuminating dendrites and spines. The second laser was directed to excite the green fluorophore in a single spine; the team could measure how fast the illuminated protein diffused into or out of the spine head.

Using this technique, Bloodgood could scan hundreds of different spines, enough to determine overall patterns of diffusion. She found that spines are highly variable, and some of them release their contents more than 100-fold more slowly than others. A subset of the spines never seemed to lose fluorescence for the two seconds that they were watched. But when observed later, some of these spines had become fast diffusers. This suggested to the researchers that the process was regulated in some way. “Presumably something’s happened in the spine and in the cell that now made it change,” Sabatini said.

When the team investigated further, “essentially what we found is that it’s regulated by activity levels,” he said. When the researchers chemically blocked excitatory synapses, which would tend to lower activity in the cell, more of the spines became fast diffusers, dumping their contents readily into the cell. When the scientists blocked inhibitory synapses, raising cell activity levels, more of the synapses held their contents back.

The spines seem to respond to global cell activity, but what about the activity of an individual synapse? With a small pipette, the team filled cells with another light-activated molecule—this time a smaller one that would mimic small signaling molecules in neurons. They first tried giving the cell an electric current to induce an action potential, but noticed no difference in the spine’s diffusion. Next they used a technique that allowed them to activate an individual synapse chemically with glutamate—still no change.

But when they combined these two signals, the spine responded. The diffusion time became slower and slower, as if the neck were clamping down. The same combination of firing an action potential while chemically stimulating the synapse is known to lead to a change in synaptic strength, dubbed long-term potentiation. “For the synapse, there are certain classes of activity patterns that are very meaningful and that trigger regulation of the synapse,” Sabatini said. “The idea is that correlated activity is telling the cell that that synapse is relevant to it.”

The current study does not prove that this shape shifting carries functional consequences for the cell, but Sabatini believes that slowing down diffusion rates could have important effects on the strength of synapses. Spines with narrow necks have a chance to accumulate signaling molecules, potentially allowing them to send stronger messages. Sabatini and Bloodgood would like to see whether the neck can even act as a resistor for electrical charge, allowing the spine to build its own small action potential.

Sabatini sees the regulation of the spine neck as an example of what has been called “metaplasticity,” the ability of synapses to change how changeable they are. When a spine narrows its channel of communication with the cell, he said, “it’s regulating its own autonomy.” Independent spines may hold more sway in the synaptic chatter of the cell.

—Courtney Humphries

top