In This Section      

Neural Connections

Brain cells appear to possess mechanisms that function as a sort of red light, stopping neurons from connecting with one another before they are ready. Flaws in these controls may lead to certain forms of intellectual disability.

auditory_hair_cells.jpg Auditory hair cells Dendrites with spines. Courtesy of Mathias De Roo through Wikimedia

May 2012—Imagine that there were a drug that could improve the intellectual function in people with autism or other cognitive disorders.

No such drug exists—and the prospect for one continues to elude us, says Seth Margolis, an assistant professor of biological chemistry. However, what’s new is that researchers like Margolis are now seriously contemplating the scientific avenues that could lead to such a drug. Neuroscientists now know so much more about the molecules and molecular pathways of cognition that such notions are no longer just pipe dreams.

Margolis’ research focuses on the formation of synapses, the communication junctures between neurons. The brain’s estimated trillions of synapses enable our every thought and gesture. But they don’t form haphazardly. “One exciting question is, what are the mechanisms that control those synapses, that determine when and where they form and how many form?” says Margolis.  In addition, “I want to understand how genetic mutations or environmental influences disrupt the number, location or timing of those events and what role these changes play in cognitive deficiency.” 

The consequence of such disruptions can actually be visualized, as Margolis demonstrates one day by calling up on his computer screen a pair of images. Each shows a magnified dendrite, one of the branching protrusions found around the cell body of a neuron. Individual neurons contain two types of projections—several dendrites, and one long protrusion called an axon. Neural communication takes place when electrochemical signals are transmitted down the axon of one neuron, across a synapse, and to the dendrite of a second neuron. 

However, a dendrite is not simply a straight, naked branch. Instead, short protrusions jut out all along its length, like the thorns on a rose stem. These protrusions are called dendritic spines, and they receive the incoming signals that traverse down an axon and across a synapse.
The two images on Margolis’ screen show this intricate morphology, but they are strikingly different. The image on the left shows a healthy dendrite. Its spines are stout, stubby and prolific, and they appear at roughly regular intervals. In the dendrite on the right, however, the spines are spindly, delicate and sparse. This dendrite came from an infant with an intellectual disability, says Margolis. He believes that such abnormalities—too few or too many spines, or spines that appear in the wrong place or at the wrong time in development—are an underlying hallmark of certain cognitive disorders, such as Down syndrome, Fragile X syndrome and a rare condition called Angelman syndrome. “What I want to know,” he says, pointing to the screen, “is what is the molecular biology of that.” 

Spines Dendrite with spines. Source: Wikimedia

Margolis hypothesizes that neurons possess a sort of “switch” that, when “flipped on,” promotes the formation of a dendritic spine—thus, a synapse. In addition, neurons contain “checkpoint mechanisms,” or “brakes,” that prevent synapses from forming prematurely. “These mechanisms tell the neuron, ‘Don’t go forward until certain events happen,’” says Margolis. He further hypothesizes that in certain disorders, the “brakes” are never released and relentlessly continue to restrain the development of dendritic spines and synapses. The result: fewer and weaker dendritic spines.

One such checkpoint mechanism appears to be a protein called EPHEXIN5, says Margolis. In addition, one “switch” for synapse creation appears to be a protein called UBE3A. Early in development, EPHEXIN5—the brakes—restricts synapse formation. Then, other molecular events ensue that activate UBE3A. Once turned “on,” UBE3A binds to and degrades EPHEXIN5, thus releasing the brakes and allowing a synapse to form. Studies in mice and cells Margolis conducted while a postdoc in the lab of neurobiologist Mike Greenberg at Harvard Medical School support this hypothesis.

But what happens if one of these molecules malfunctions? Then the brakes might not be released, and a synapse would not form.

Margolis suspects that such a malfunction occurs in Angelman syndrome which occurs in an estimated 1 out of 15,000 people worldwide and is characterized by developmental delays, speech impairments, seizures and problems with walking and balance. People with Angelman syndrome also have a tendency to laugh and smile frequently.

In most cases, Angelman syndrome involves a mutation in the UBE3A gene. This mutation could prevent the UBE3A protein from degrading EPHEXIN5, says Margolis. Using a mouse model of the syndrome, Margolis is now dissecting the interaction of these two molecules. Armed with that knowledge, he hopes to be able to design small-molecule drugs that compensate for the defective UBE3A protein. For example, one plan would be to develop a small molecule that inhibits EPHEXIN5, in effect releasing the brakes without UBE3A and permitting new synapses to sprout.

It’s unlikely that such a drug would cure Angelman syndrome, says Margolis. But it might ameliorate certain features, perhaps reducing the risk of seizures or improving the quality of an affected person’s speech. “Angelman syndrome could be one of the first cognitive disorders that could be treated.” Such a possibility, says Margolis, “drives me forward.”

 —Melissa Hendricks

Related Stories:
Seth Margolis on never underestimating the learning potential of a failed experiment