The paper “seems to be a very thorough study that addressed basic mechanisms underlying the regulation of dendritic spines from many different angles, including using some very exciting new methodology,” said Shelley Halpain, of the University of California in San Diego. “I think that basic research studies like this one are going to play a very important role in helping researchers interested in Alzheimer’s and other diseases approach their studies.” Halpain was not involved in the current work.

In muscle cells, myosin II provides contractile force by pulling on the actin cytoskeleton, while in other cell types, the myosin-actin interaction regulates cell growth. In the axon growth cone, myosin has been shown to shear apart actin bundles, producing the monomeric building blocks needed to extend actin filaments and elongate the axon (see Medeiros et al., 2006; Vallee et al., 2009). Actin filaments give structure to dendritic spines as well, and changes in actin structures are known to mediate synaptic plasticity. Myosin II activity has been found to be necessary for the development of dendritic spines in cultured neurons (see Zhang et al., 2005 and ARF related news story on Ryu et al., 2006), but little was known about the motor protein’s role at mature synapses and in synaptic plasticity.

To investigate this, first author Christopher Rex and colleagues used short hairpin RNAs (shRNAs) to knock down myosin II expression in adult rat hippocampus. Normal synaptic activity was not changed in the knockdowns, but LTP was impaired, failing to stabilize after induction. The same results were obtained with blebbistatin, an inhibitor of myosin II activity. The authors also showed that the first 10 minutes after induction is the critical window for LTP stabilization, with myosin activity required in the first 30 seconds. Other experiments demonstrated that LTP induction activates myosin through the N-methyl-D-aspartic acid (NMDA) receptor and downstream Rho-GTPase signaling.

To see what was happening to actin at the synapses during this time, Rex and colleagues made use of a cutting-edge technique in which acute hippocampal slices are bathed in fluorescently labeled phalloidin, which diffuses into neurons and binds actin filaments, allowing them to be seen under the microscope (see Lin et al., 2005). They found that new actin filaments normally appear in dendritic spines two minutes after LTP induction, and that when myosin II activity is blocked, these filaments don’t form. The results indicate that motor protein activity is necessary for actin polymerization at synapses, and also that myosin activation, because it is required as early as 30 seconds, precedes filament formation. Intriguingly, Rex and colleagues found that inhibition of myosin II led to less turnover of actin filaments, suggesting that myosin II acts at the synapse just as it does in growth cones—by disassembling old actin structures and allowing new ones to be built.

To tie these synaptic plasticity effects to memory formation, the authors inhibited myosin activity in adult rats by injecting shRNAs to knock down the protein. In fear conditioning trials, treated rats learned the context-shock association normally, but showed less freezing behavior 24 hours later compared to mock-injected rats. This suggests that myosin is required for long-term memory storage.

The results support the hypothesis that a destabilization of the actin cytoskeleton allows the synapse to be rebuilt in a new way to store information, said Rumbaugh, and also show that myosin II is a key mediator of this process. “It rips apart the old synapse and then builds a new one,” he said, but added that myosin probably does not directly assemble monomers into new filaments. To pin down the exact role of the motor protein, Rumbaugh said, his lab is developing tools to track myosin motor activity by following changes in a fluorescent reporter molecule. Theoretically, this system could measure myosin dynamics at the millisecond timescale, allowing researchers to build a detailed picture of synaptic events occurring after LTP induction.

These findings also have implications for Alzheimer research and therapies. Aβ has been shown to disrupt actin dynamics at the synapse (see ARF related news story on Zhao et al., 2006; Salminen et al., 2008), and recent studies have implicated myosin II in APP processing (see Argellati et al., 2009), suggesting that these pathways could become drug targets. Halpain points out that “destabilization and weakening of synaptic connections is something that precedes the overt loss of neurons [in AD]. Now, a newfound focus on mechanisms specific to synapses is going to change how we think about the disease process and how we might approach it therapeutically.”

Rumbaugh says his group is interested in “finding a way to pharmacologically activate myosin motors, because we think that that would facilitate learning and memory.” Because NMDA receptors are found only in the brain, Rumbaugh speculates that other elements of the synaptic myosin activation pathway may be specific to the brain, which would allow development of drugs that target this pathway without causing side effects elsewhere in the body. His group is also studying how actin dynamics change during development. “We suspect that slowing of actin dynamics at synapses is part of the aging process. We are testing the idea that re-instating rapid actin dynamics in the mature brain will reorganize brain circuits and enhance memory and cognition.” This would be important in Alzheimer’s, where synapses are failing.—Madolyn Bowman Rogers.

Reference:
Rex CS, Gavin CF, Rubio MD, Kramar EA, Chen LY, Jia Y, Huganir RL, Muzyczka N, Gall CM, Miller CA, Lynch G, Rumbaugh G. Myosin Iib regulates actin dynamics during synaptic plasticity and memory formation. Neuron. 2010 Aug 26;67(4):603-17. Abstract