"Andrew Gyles" <syzygium at alphalink.com.au> wrote in message news:<9p37k3$g4h2n$1 at ID-94640.news.dfncis.de>...
> "Matt Jones" <jonesmat at physiology.wisc.edu> wrote in message
> news:b86268d4.0109281042.6185f69a at posting.google.com...> > "Andrew Gyles" <syzygium at alphalink.com.au> wrote in message
> news:<9om3lc$e05sl$1 at ID-94640.news.dfncis.de>...
> Thank you for your comments. All of the questions you posed are valid and
> would have to be answered by experiment before my hypothesis was accepted or
> rejected. Perhaps I can attempt a preliminary answer to a couple of them.
>> Transcriptional regulation of genes for K channels would be a slower form of
> "switching" than the "flip-flop" action of mitochondria that I postulated,
> would it not, and speed is important in information processing. Perhaps the
> "expression of previously silent AMPA synapses" would also take longer than
> the action I suggested.
>> On the other hand, latching of CAM kinase II into an autophosphorylated
> state would perhaps be quicker than the "flipping" or "flopping" of a
> mitochondrion. And as you remarked there are other alternative possible
> switching actions. My hypothesis is just one of many and will have to face
> the test of experiment.
>> Andrew Gyles
I can't recall the details of your mitochondrion hypothesis. But yes,
you are right, all the things I suggested would probably be slower
than the mechanism you proposed.
Transcriptional regulation would be quite slow, taking hours or even
The autophosphorylation and the silent synapses ones could be as quick
as a few seconds, because both have been proposed as mechanisms
underlying LTP (long-term potentiation), which can be induced fairly
However, if flip-flopping on the order of tens or hundreds of
milliseconds is necessary, then something faster would be needed. I
think a "network" solution could potentially operate on this fast
timescale. For example, suppose you had an excitatory neuron that
either synapsed back onto itself (this has been observed in cultures
and slices) or excited itself via an intermediate excitatory neuron
(this also has been observed). Then a spike in this neuron would lead
to an EPSP, which could trigger another spike, etc. Such a simple
circuit would act as a digital "latch", and keep itself on after being
activated. This probably has rather different implications than your
theory, and obviously, requires external units. Your theory has the
nice feature that a single neuron could potentially implement multiple
flip-flop elements at the same time.
I don't actually remember the details of your theory, but I also seem
to recall being concerned with some of the biophysical details
included, I think related to charge storage or something like that.
By the way, are you also aware that mitochondria can store and release
intracellular calcium in response to certain stimuli? This could
potentialy be incorporated into a flip-flop theory of mitochondrial
It's interesting to think about these things. But i hope you will not
mind if I remain skeptical for now.