Matters of Size

See allHide authors and affiliations

Science  08 Dec 2006:
Vol. 314, Issue 5805, pp. 1554-1555
DOI: 10.1126/science.1137595

From the overall body plan of an organism to the intricate three-dimensional fold of proteins, structure is a key determinant of function. Neurons, the fundamental cells of the nervous system, are no exception. The architecture of their dendritic and axonal arbors—the cellular extensions that receive and transmit information—determines which neurons they can connect to, whereas the diameter of these extensions determines the speed and filtering of electrical signals that travel down them. Tiny femtoliter (10−15 liter)-sized protrusions from neuronal dendrites, called spines, receive a functional connection from another neuron's axon at a specialized area of contact known as a synapse. A study by Park et al. in a recent issue of Neuron (1) marks a large step forward in our understanding of how spine size and synaptic strength are balanced.

A neuron can have up to 100,000 spines, each generally forming a single synapse. Spines function as chemical compartments for signaling molecules that become activated by specific patterns of synaptic transmission (24). This organization provides each synapse with a miniature caldron in which to concoct a chemical brew to effect changes in connections between neurons (5).

Interestingly, large spines contain strong synapses (robust transmission) and small spines have weak synapses (6, 7). A spine is at least an order of magnitude larger than a synapse, and thus there is no physical requirement for this correlation. The reason for this correlation between structure and function remains elusive, but an abundance of circumstantial evidence points to its importance. Stimuli that cause stable changes in synaptic strength lead to corresponding stable changes in spine volume (8, 9). Heritable forms of mental retardation can present abnormalities in spine morphology as well as synaptic function (10). Furthermore, Alzheimer's disease may involve a loss of spines that is fundamentally linked to a decrease in the number of neurotransmitter receptors at the synapse (11). Therefore, understanding how and why this correlation between synapse strength and spine size exists will not only expand our understanding of how synapses work, but may have clinical relevance as well.

Park et al. elegantly combine serial section electron microscopy and live cell fluorescence microscopy to afford us a view of the inner workings of spines. The authors stimulated cultured mammalian neurons to generate a stable increase in synaptic strength known as long-term potentiation (LTP), and confirmed that the rapid increase in synaptic strength is accompanied by a matched increase in spine volume. They then probed the molecular and cellular mechanisms behind this correlation.

Park et al. focused on the role of the recycling endosome, an intracellular membrane-bound compartment that is part of the system that transports membrane-bound proteins onto and off the cell surface. Previous work by this group showed that the protein GluR1 is delivered to the neuronal surface from the recycling endosome through exocytosis, the cell's secretory process (12). GluR1 is a glutamate receptor subunit that is inserted into synapses during LTP and plays an important role in mediating the increase in synaptic strength (13). Blocking this delivery by expressing mutant proteins that specifically inhibit this exocytosis prevented the stable increase in synaptic strength.

In the present work, Park et al. provide tantalizing evidence that the lipids delivered to the neuron's surface from the vesicles carrying GluR1 are the raw materials that allow the spine to enlarge (see the figure). The recycling endosome appears to be situated in the right place, just below or even within some spines, and is of sufficient size to influence spine volume. LTP-inducing stimuli mobilize these endosomes from dendrites into spines, positioning the endosome perfectly to fuse with the spine surface. Blocking exocytosis from this compartment prevents spines from enlarging, strongly suggesting that the recycling endosome is a source of structural plasticity. Furthermore, the amount of surface area lost in the endosomal system equals the amount gained by the spines, hinting at a direct transfer of material. Park et al. also directly visualize exocytosis with a pHsensitive fluorescence indicator that translates the pH change experienced during exocytosis (the pH inside the recycling endosome is acidic, whereas in the extracellular space it is mildly alkaline) into a large change in fluorescence. By monitoring events simultaneously, these experiments reveal that exocytosis takes place directly in spines and that the amount of exocytosis correlates extremely well with the increase in spine volume.

Balancing act.

Long-term potentiation drives exocytosis of recycling endosomes, providing dendritic spines with more membrane and receptors (GluR1). Actin polymerization provides structural support. These processes are somehow balanced to regulate the size of spines and the strength of synaptic connections.


Although this study elucidates how spine size and synaptic strength are kept in check, it is not the whole story. Several groups have investigated the role of the actin cytoskeleton in determining spine morphology (14, 15). Indeed, LTP causes an increase in the amount of filamentous actin in spines (16, 17), and preventing the formation of filamentous actin blocks structural (16) and functional (18, 19) changes during LTP. It is difficult to imagine how lipids that are added to the spine membrane could be sufficient to make a larger spine, rather than simply flow off into the membrane of the dendrite. It is thus likely a combination of actin polymerization and the exocytosis of recycling endosomes that mediate spine enlargement during LTP. Filamentous actin acts as a skeleton to support a larger spine, whereas more lipids are the raw material to increase the spine's surface area.

But if these two processes are required for structural and functional plasticity, how are they balanced? That is, how are the distinct molecular cascades underlying exocytosis and actin cytoskeletal reorganization coordinated? Perhaps evolution has perfectly balanced their rates, or maybe there is a physical link between the two systems. For instance, receptors delivered to the synapse from the recycling endosomes could stabilize the actin cytoskeleton and thereby provide a simple accounting process to balance changes in synaptic strength and spine size. Maybe when we fully understand how spine size and synapse strength are coordinated will we be poised to comprehend why spine size matters.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
View Abstract

Stay Connected to Science

Navigate This Article