glenne at csd4.csd.uwm.edu (Neuromancer) writes:
:: Perhaps you remember the article posted on this newsgroup regarding
: the use of tree grammars to model membrane potentials. I showed the
: article to my formal languages/grammars/automata class professor, and
: he'd like me to discuss it with the class.
:: I have gotten a rough idea of what tree grammars are and would like to
: get some confirmation and correction of my views of membrane potentials.
: I am a computer scientist at this point, and have picked up a very limited
: amount of neurobiology by reading neuroscience books for entertainment.
: I wish to clear up some of my "knowledge" before I email questions to the
: person who posted the article.
:: Most of this is pure guesswork. I am assuming that there is such a
: thing as a dendritic bouton.
Actually it is not usually called a dendritic bouton, but a dendritic
spine. The term bouton is the specxialization on the cell that is
releasing the neurotransmitter onto the dendritic spine, and it is usually
called a synaptic bouton, or a presynaptic bouton.
Suppose we are at the dendritic bouton. Some
: neurotransmitters are released onto this bouton causing an EPSP. All
: EPSP's and IPSP's die out due to the same reason that we have a resting
: potential. (The membrane at this bouton allows some of the positive ion's
: out while not allowing other positive ions in.)
Actually it is a little more complicated than this. Depending on what
receptors a re present in the dendritic spine, different ions may flow
onto or out of the cell through certai channels, usually sodium and
calcium in, and potassium out. This depolarizes the cell locally, causing
the local generation of the thing called the EPSP. The EPSP is conducted
passively away from the site of initiation, but soon decays to zero.
However it is large enough (ie exceeds some threshold value) a different
class of iion channels will be opened (voltage sensitive ions channels)
that allow sodium in and potassium out. To cut a long story short, this
causes the generatio of what is called the action potential, which is self
propagating and will travel to the end of the cell provideing the density
of voltage sensitve ion channels is sufficient along the way.
I would guess that some
: of the voltage of these EPSP's will overflow onto other nearby boutons and
: would still be considered EPSP's as long as they died out.
:: However, suppose some more neurotransmitters are released on this bouton.
: only this time, many EPSP's are occuring on all kinds of nearby
: dendritic boutons. In fact there are so many EPSP's occuring that the
: potential cannot be drained away through this bouton's membrane, nor the
: nearby membranes, but instead the voltage flows all the way to the soma
: and down the axon where the electrical potential is converted the chemical
: ejection of neurotransmitters. I.E. after a threshold amount of EPSP's
: occur an action potential occurs. (Is an action potential synonymous
: with a spike?)
Yes , action potential is synonomous with "spike".
: However while this would seem to be a reasonable explanation of how
: ESPS's get translated into an action potential, it might be even more
: efficient if somehow, once an action potential had built up,
: this would change the neural membrane so that the action potential could
: not leak out at all through the membrane as it traveled to the soma and
: the axon.
This is more or less the case in myelinated axons. The ion channel density
is not high enough throughout the entire axons, but focal concentrations
occur at what are called nodes of Ranvier. There, the action potential is
recreated if the current conducting in to that region fro the last node is
enough to depolarize the axon there beyond threshold, opening the voltage
gated channels at the new node. This fires off a new action potential,
which then conducts pasively (but very rapidly and slowly decaying) on
down to the next node and so on. So you could say that myelinated axons
have evolved somewhat along the lines of how you were speculating. I
recommend that you take a look at the chaptersin Kandel, Schwartz and
Jessell (7-10) that deal with this.
:: Anyone care to shed any light on this wild speculation?
:: Thanks in advance,
:: Glenn glenne at csd4.csd.uwm.edu