Spring 1995 Volume 4, Number 2

The Inner Life of Neurons

An empire of semi-independent domains

BY MARINA CHICUREL, PH.D.
AND CHRISTOPHER DEFRANCO, PH.D.

Ultimately, the key to understanding learning and memory lies in the study of the properties of neurons - and particularly in their interconnections, called synapses, which are the sites at which nerve cells communicate.

Neurons are large cells with numerous, widespread branch-like projections (referred to as neurites) that extend many microns and sometimes as much as a meter (30 inches) from the cell body. Where the long neurite, known as an axon, extending from one nerve cell meets the shorter, stubbier neurite, called a dendrite, of another nerve cell, a synapse is formed and the stage is set for the two cells to communicate.

A remarkable feature of neuronal cells is the very large number of synapses they establish and maintain. A typical nerve cell in the human brain, for example, harbors tens of thousands of synapses, many of which may be communicating at one time. Perhaps even more remarkable is the fact that these contacts are not fixed, but are constantly changing in a regulated manner throughout our, or any animal's, lifetime. It is precisely these changes that appear to underlie learning and memory.

The thousands of synapses established by a single neuron, moreover, do not function as a single entity, but rather are modulated, at least in part, independently of one another. For example, it appears that during the process of learning, some of the synapses on a single cell are strengthened (the efficiency of neurotransmission is increased) while others remain unchanged. Neuron Photograph

The neuron. The orange dots represent the multiple synapses on a single neuron. (Photo courtesy of Drs. P. CeCamilli, G. Banker, T. Sudhof and M. Matteoli, copyright held by Cell Press.)

If you envision trying to answer the phone and the doorbell while fixing coffee for a visitor with whom you are having an important conversation, you have a very simplistic idea of how complex the mechanism must be by which the enormous number of separate synapses on a single cell is properly maintained and regulated. As you spill the coffee and miss the call, you can see how hard it is to imagine how a neuron can achieve local and independent control of its multiple synapses through centralized management alone.

We and others have obtained evidence indicating that one way in which this managerial feat may be accomplished is by locally regulating the manufacture of the primary agents of cell function - the proteins - at individual synapses. Specifically, we discovered in proximity to synapses in the brain the two key components required for protein synthesis. This is unexpected, as we are accustomed to seeing the molecules that control gene expression located at some central point within the cell.

The two components are the manufacturing units, called ribosomes, and the "templates" that encode the manufacturing instructions, the mRNA molecules. We also found that the mRNA molecules present at synapses constitute a restricted population that is distinct from the RNA population found in the cell body. The synaptic mRNA population includes the templates for proteins needed for the specialized function of the synapse.

These findings suggest that neuronal proteins can be made locally in response to the signalling activity occuring at each individual synapse. Thus, proteins can be supplied in near proximity and quickly, according to the particular needs of each of these independent domains of the cell - just as an overseas manufacturer might serve foreign customers by setting up a local assembly plant.

But turning up these crucial protein fabrication components at the synapse only half-answers the question of the neuron's managerial virtuosity: How are mRNAs transported to the synapse? How are they anchored in place once they get there? Does, indeed, synaptic activity regulate protein synthesis at the synapse? To begin answering these questions, we are studying how mRNA interacts with other cellular components.

It has been known for some time that most of the mRNA in a cell is attached to a network known as the cytoskeleton. The cytoskeleton provides cells with their characteristic shapes and forms, as the bones in the human body do. In addition, one of the major jobs of the cytoskeleton is to provide a highway system to transport cellular components to different locations and compartments within the cell.

Researchers have recently discovered that mRNA molecules appear to be among the cellular components that are transported by the cytoskeleton. Our current work has focused on investigating how mRNA is attached to the cytoskeleton, what regions of the mRNA are important for this attachment, and what proteins mediate the interaction.

We have discovered that most mRNA in cells appears to be contained in cytoskeletal-linked particles, which we have named "cytosomes." The cytosome may create an RNA-protein scaffold important in the transport of mRNAs and in the regulation of protein synthesis.

Neurons are but one kind of cell that can benefit from making proteins locally. Many other cell types, such as an unfertilized egg (or oocyte), also use mRNA localization to direct local events within the cell. This strategy is cost effective in terms of energy and organization: It is far easier to move the template than it is to move all of the products made from that template. In addition, having the template at the site of need, prepared for producing proteins in direct response to local incoming signals, bypasses the problem of long-distance relay of signals and proteins throughout the cell.

We have come to understand the tissues of higher organisms as collections of cells whose individual behavior determines the maintenance and regulation of the tissues' varied functions. But just as a tissue is composed of multiple cells, cells are composed of multiple compartments. Thus the neuron can be viewed as an empire of numerous domains, whose independent function may ultimately determine how higher brain processes such as learning and memory occur.*

Dr. DeFranco is an instructor and Dr. Chicurel is a postdoctoral fellow working with Dr. Huntington Potter in the Department of Neurobiology at Harvard Medical School.

Table of Contents