Thanks again everybody
That helps me much !
Laurent.
Matt Jones wrote:
> I do not think this waiting is necessary or desireable. Most of the
> time, the nervous system attempts to do everything as fast as it can,
> given the constraints of its architacture, and the constraints imposed
> by physics. Typically, it achieves near optimal speed given the
> materials it is made of.
>> However, these constraints -do- introduce delays. Here is a list of
> approximate delay times associated with various steps in transmission
> between two run-of-the-mill neurons in cortx or hippocampus:
>> 1) Delay from the time the spike (action potential) is fired at the
> cell body until the propagating spike reaches the nerve terminal -
> about 1-3 msec.
>> 2) Delay from the time the spike reaches the terminal until calcium
> channels are fully open - about 100 usec (microsec).
>> 3) NOTE: although calcium channels are open at the peak of the spike,
> not very much calcium will enter until the spike starts to fall. This
> is because the peak of the spike is close to the calcium reversal
> potential (see a good neuroscience textbook if you don't understand
> what I just said). Therefore -most- calcium entry occurs on the
> falling phase of the spike (perhaps it was something like this that
> you read about that made you think the spike needed to be over before
> transmitter was released). So, the delay from the initial opening of
> Ca channels until the bulk of calcium enters is about the duration of
> the spike - about 0.5-1 msec.
>> 4) Time required for activation of the release machinery and fusion of
> the vesicle - about 100 usec. NOTE: there are also very delayed
> components of release that can occur up to several hundred msec after
> calcium entry, but the main portion happens very quickly.
>> 5) Time required for diffusion of neurotransmitter out of the vesicle
> and across the synaptic cleft - about 10 usec (that's right, ten
> microseconds). As Richard mentioned, diffusion is extremely fast. It
> has to be, since its essentially the mechanism underlying evry other
> motion in biology. Basically, nothing much can be faster than
> diffusion.
>> 6) Time required for binding to, and initial activation of
> postsynaptic receptors - about 10-25 usec.
>> 7) Time required for -PEAK- postsynaptic response - if the receptors
> are ligand-gated ion channels, their peak opening will occur within
> about 0.5-10 msec. If the receptors are G-protein-coulpled, the peak
> response will occur about 100 msec later (because of multiple
> biochemical events between receptor activation and the final
> response).
>> 8) Time required for the decay of the postsynaptic response - for
> ligand-gated channels, the currents will decay in 1-200 msec,
> depending on channel type. For g-protein coupled responses, usually
> last for about 0.5-2 seconds. Also, for really fast ligand-gated
> responses, the ultimate duration of the postsynaptic potential change
> will be roughly the membrane time constant - about 10-30 msec.
>> So, the total time from spike firing at the cell body until the peak
> of the postsynaptic response is typically on the order of 1-4 msec.
> The largest delay by far is the time it takes the spike to propagate
> down the axon (which depends on axon length, of course).
> Interestingly, the different delays from different axons can be used
> to establish timing mechanisms, called delay lines. This is used in
> the auditory system to locate sounds horizontally, by exploiting the
> interaural time-diffeence (i.e., time between when the sound reaches
> one ear, then the other).
>> Cheers,
>> Matt
>> Cheers,
>> Matt
