LTP and excitotoxicity

Bill Skaggs bill at
Tue Oct 25 15:21:24 EST 1994

william at (Fiberman) writes:

   > What I like to know is this.  Is the stimulating parameters one
   > uses to induce LTP (like high frequency tetanus) occurring
   > naturally?  

Possibly, in either of two ways:

1. There are events in the normal hippocampal formation, called "sharp
waves", that resemble 100-200 msec long epileptic seizures.  In a
single sharp wave, 10 to 20 percent of the cells in the CA1 region
fire, often more than one spike.  Sharp waves occur once every second
or two whenever animals are resting or in non-dreaming sleep.  Georgy
Buzsaki argues that the input seen by certain parts of the hippocampal
formation during sharp waves is very similar to the sorts of stimuli
normally used to induce LTP.

2. There is another form of LTP, usually called "theta burst
potentiation", that can be induced by a very short train of stimuli,
if they follow the right temporal pattern, which is modeled after the
"theta rhythm" seen in hippocampal EEG in many species.  (This was
originally discovered by Greg Rose.)

And there is another important point.  LTP can be induced in many
parts of the brain if inhibition is reduced by GABA antagonists.  The
reason why it can be induced in the hippocampus without any such
assistance, is probably that the hippocampus has an especially low
fraction of inhibitory interneurons.  (This is also the reason why the
hippocampus is particularly susceptible to epilepsy.)  Consequently,
as one increases the intensity of stimulation, the evoked inhibition
saturates long before the evoked excitation has maxed out, i.e., with
strong stimulation, excitation overwhelms inhibition.  This seems to
be the essential requirement for LTP -- but under natural conditions
it may be achievable in several ways other than using high frequency
trains of strong stimuli, for example by suppressing the interneurons
via neuromodulatory inputs from other parts of the brain.  

Intracellular studies tell us that the crucial requirement for LTP is
presynaptic activity followed very shortly by strong postsynaptic
depolarization, and it seems almost obvious that this will occur
often under natural conditions.

   > Alternatively, if LTP occurs naturally in an animal, one should
   > able to see it in a slice without artificial stimulation.  

I don't think there's any way to see "natural" LTP in a slice, because
slices are very unnatural.  To observe LTP you would need to measure
synaptic strength (by whatever method), induce LTP (by whatever
method), and then measure synaptic strength again.  It's relatively
easy to measure synaptic strength in a slice, but anything you do to
induce LTP in a slice is by definition unnatural (merely because it's
occurring in a slice).

In principle the same thing could be done in a living animal, but
alas, it requires intracellular recording, and nobody has yet
succeeded in recording intracellularly from freely moving animals.
With the techniques that are currently available, circumstantial
evidence seems to be the best that is possible.

	-- Bill

More information about the Neur-sci mailing list