Published: Oct. 29, 2003
Updated: Nov. 3, 2004
Reporters & producers can visit Duke Medicine News and Communications for contact information.
By Duke Medicine News and Communications
DURHAM, N.C. -- Duke University Medical Center neurobiologists have discovered how neurons in the brain "reset" when they are overly active. This molecular reset switch works to increase or decrease the sensitivity of brain cells to stimulation by their neighbors. Such "homeostatic plasticity" is critical for the brain to adapt to changes in the environment -- either to avoid having its neurons swamped by increased activity of a neural pathway, or rendered too insensitive to detect triggering impulses from other neurons when neural activity is low. This plasticity is distinct from the more rapid changes in neural circuits laid down early during the formation of memories, said the scientists.
According to the researchers, their basic studies provide long-sought clues to how neurons protect themselves during stroke, epilepsy, and spinal cord injury. Also, their findings may help explain diverse brain changes that occur during early childhood and that go awry in later stages of life in Alzheimer's or Parkinson's disease.
The researchers, led by Assistant Professor of Neurobiology Michael Ehlers, M.D., published their findings in the Oct. 30, 2003, issue of the journal Neuron. Other authors are Yuanyue Mu, Ph.D., Takeshi Otsuka, Ph.D., April Horton and Derek Scott. The research was supported by the National Institutes of Health, the American Heart Association and a Broad Scholars Award.
According to Ehlers, homeostatic plasticity has long been theorized to exist and recently demonstrated in mammalian neurons, but no mechanism for the process had been discovered. Neuroscientists believed that such plasticity somehow adjusted the global sensitivity of the myriad connections, called synapses, that a neuron makes to its neighbors. At each of these synaptic connections, one neuron triggers the firing of a nerve impulse in another by launching a burst of chemicals called neurotransmitters across a gap called the synaptic cleft.
Such homeostatic plasticity takes place over periods of hours or days and adjusts all a neuron's synapses, said Ehlers. In contrast, the synaptic adjustments that underlie rapid learning take place within seconds.
"Neurobiologists have understood that a neuron can increase only so much its firing rate in response to inputs from other neurons, and then it saturates," said Ehlers. "There had to be a way for a neuron to recalibrate -- to scale up or down to stay within an optimal dynamic firing frequency range.
"Consider when you're driving a car with a manual transmission. As you accelerate, you reach a point where the engine's RPMs are maximal and can go no higher. At that point, you need to switch gears to bring back your RPMs to an optimal range. What we have found is the molecular clutch that allows neurons to shift gears" said Ehlers. "This really is a profoundly important discovery. Imagine if your brain could only operate in 'second gear,'" he said.
Although homeostatic plasticity had been a theory, only in the last couple of years has its existence been confirmed functionally, said Ehlers, and its mechanism was a mystery. It was known that the process involved changing the number of neurotransmitter receptors -- the proteins on the surfaces of synapses that serve as receiving stations for neurotransmitters. One key type of receptor implicated in such changes is called the NMDA receptor -- a major component of molecular learning and memory. The level of NMDA receptors was known to increase or decrease over time periods consistent with homeostatic plasticity, said Ehlers.
Using an array of analytical techniques, Ehlers and his colleagues showed that the level of NMDA receptors was controlled throughout a neuron by the processing of its initial genetic blueprint -- called messenger RNA (mRNA). Messenger RNA is a copy of the genetic DNA blueprint for a specific protein -- such as the NMDA receptor protein -- that the cell's protein-making machinery uses to manufacture the protein.
Specifically, the researchers' experiments revealed that when a neuron needs to increase its overall sensitivity, the stringlike mRNA blueprint for the NMDA receptor is snipped apart and spliced back together slightly differently than when the neuron needs to decrease its sensitivity.
This "alternate splicing" causes the production of an NMDA protein differing in one small bit that attaches it to the transport machinery that carries receptors to the synaptic surface, found the Duke neurobiologists. When the overall triggering of a neuron decreases, and the neuron needs greater sensitivity -- and thus more NMDA receptors -- the alternate mRNA splicing yields a receptor protein with a variant of this bit that encourages the receptor to attach to the transport machinery.
"It's a very surprising mechanism, and it explains a lot," said Ehlers. "It explains why the process is relatively slow, because the process of changing splicing of new proteins would be slow. And it explains how this process can happen at all the synapses on a neuron, because it happens in the neuron's nucleus, where mRNA splicing happens."
According to Ehlers, the findings by him and his colleagues could aid understanding of how brain tissue is damaged during stroke, and altered in pathological states of addiction or following injury.
"For example, it's been known for some time that the circuitry of the spinal cord is altered in response to spinal cord injury, enhancing the NMDA receptor-mediated transmission of nerve impulses," said Ehlers. "This aberrant rewiring causes all kinds of problems in patients, including heart arrhythmias and hypertension," he said. "So, our studies could lead to new therapeutic approaches for treating such problems by targeting the alternate splicing of mRNA for NMDA receptors."
More broadly, said Ehlers, "these findings could open a floodgate of studies to determine where else in the brain alternate splicing is used as a central control mechanism. It is known that the brain uses alternate splicing more than any other organ, but until now there has not been an experimental system in which a specific alternate splicing event could be controlled and studied.
"We have identified a completely new cellular signaling pathway, and it's going to be quite exciting to unravel how it works," said Ehlers. "Potentially this could open a whole new window into a very fundamental aspect of neuronal function."