This paper reflects the research and thoughts of a student at the time the paper was written for a course at Bryn Mawr College. Like other materials on Serendip, it is not intended to be "authoritative" but rather to help others further develop their own explorations. Web links were active as of the time the paper was posted but are not updated.
2000 First Web Report
When a neuron is first formed it does not yet have dendrites, and therefore also does not have dendritic spines. Dendritic filopodia are formed from the dendrites first and then convert into spines after being innervated by synaptic fibers. Spines on different types of neurons attain their peak actin density at different times. Fewer spines are present in adults than children and there is a peak growth time during post-natal development. Adult brains show up to 50% fewer spines than developing brains (2) . Brain disorders, such as strokes, epilepsy, and forms of mental retardation like Fragile X, have been connected to deformations of dendritic spines or the total absence of them on certain neurons. Spines are predominantly found at excitatory synapses where inputs from many areas of the brain arrive.
Initially during spine formation N-methyl-D-aspartate (NMDA) is the main growth and development regulator. NMDA is a glutamate receptor found at excitatory synapses in most neurons in the mammalian brain. It contains channels permeable to calcium ions. Ions can accumulate and initiate currents at the head of the spine where the calcium channels are located, separate from the shaft of the dendrite. Weak calcium-induced currents affect individual spines whereas stronger currents can summate to affect multiple spines as well as areas of the dendrite's shaft. LTP is a strengthening of the synaptic connections which occurs when spines are formed.
There are several steps to achieve a current in a spine. Magnesium ions block the NMDA receptor sites, but are displaced when a stimulus, such as caffeine, depolarizes the receptor. Calcium ions are then able to pass through and collect within the spine. Once the amount of ions reaches a threshold level the LTP is generated. A very strong stimulus will depolarize receptors on more than one spine that can lead to the current summation effect previously described. The LTP results in increased spine motility that has been directly linked to actin in the heads of the spines. The process of LTP generation has been proved experimentally by adding NMDA receptor antagonists to the cell showing that blocking the receptors will not interfere with signal transmission at the synapse, but a LTP will not be able to form (3) .
Once spines are formed they must be stabilized and maintained. Another form of glutamate receptor, -amino-2-hydroxy-5-methyl-4-isoxazole propionate (AMPA), which is also controlled by voltage-gated calcium channels, is responsible for this stage in spine development. When AMPA receptors are activated, the actin found in protrusions in the spine's head is blocked. The actin and thus the entire head of the spine, collapses while more stable filaments in the center of the spine remain. Studies performed using NMDA receptor antagonists demonstrated that AMPA receptors will begin to inhibit or block spine motility without the presence of NMDA. However, in developing neurons when the AMPA is washed out and the NMDA remains lost spine motility is able to recover (4) .
Interestingly, a stimulus that is of certain strength can result in the maximum extension of a dendritic spine, reducing the connection between spine and dendrite. This leads to lower synaptic efficiency. However, if the AMPA is able to reduce the spine length the synaptic connection may be strengthened. A stimulus that is too great will destroy the spine completely. Unexpectedly, less activation in an adult brain, such as when synaptic activity is inhibited or blocked, can result in the growth of more spines, rather than a reduction in spines.
Many studies have been performed which link dendritic spines in the hippocampus to learning and memory. These experiments along with the general development of the spines give one more piece of evidence to the argument that brain equals behavior. If each spine is able to receive and store inputs, and there are thousands of spines on each dendrite, then it is understandable that humans can learn and retain facts and ideas throughout their lives as well as synthesize original thoughts. One new question that arose while investigating the dendritic spines is how do they affect the I-function? Since they are so vast in number, are directly associated with synaptic connections, and are found mostly on excitatory synapses it is reasonable to think that they could be linked to complicated ideas such as consciousness and being. If brain does equal behavior, then dendritic spines, being a part of the brain, must affect behavior in some way. How spines and behavior are related would be a very interesting thing to know.
2) Journal article titled "Rapid plasticity of dendritic spine: hints to possible functions?"
3) Series of movies about how NMDA and AMPA affect dendritic spines, from MIT webpage.
3) Journal article reviewing current research concerning LTP, from Nature magazine.
| Course Home Page
| Forum | Brain and
Behavior | Serendip Home |