From: Dennis Kaetzel
To: Kevan Hashemi
Subject: Re: Pyramidal Neurons
X-BtMT: Tue, 29 May 2012 15:33:11 -0400 (EDT)

Hi Kevan,
in that picture (c), the black strokes are the dendrites. the grey area 
is the soma, and the red bit is the beginning of the axon. The dendrite 
is the input compartment and the axon the output compartment - the soma 
the integrating part in between those two compartments. Essentially we 
are looking at an AD-Converter here - "random" inputs from thousands of 
synapses sitting on the dendrite and each causing a small current and 
thus change in membrane potential (making up an analog signal) are 
converted into a kind-of digital signal namely a series of action 
potentials, which are all-or-none events, that are guided through the 
axon into its terminals, the presynapse, contributing to an analog 
signal in the next neuron.
There are no synapses on the axon (with few exceptions, and those are 
mainly inhibitory synapses; or there are synapses directly on synapses, 
but lets ignore this case, too, those ones are mainly of regulatory 
nature). The dendritic tree (all dendrites) receives all the inputs via 
synapses (where currents flow through channels opened by 
neurotransmitters, and those currents causes a change in membrane 
potential). Synapses are simply the connection points between two 
neurons - it means that a terminal piece of an axon of an upstream 
neuron (presynapse) connects to some area of dendrite on the downstream 
neuron (postsynapse) (i.e. a synapse is formed by a pre- and a 
postsynapse - the presynapse releases neurotransmitter, which opens ion 
channels in the postsynapse). Those inputs flow towards the soma, where 
they are integrated, and if the excitatory ones - summed up in a certain 
window of time - are large enough to drive the membrane potential 
(interior potential relative to outside the cell, which is considered 
0mV) from ~ -70mV to -40mV, then voltage-gated sodium channels will open 
and in a positive feedback loop that unfolds in ~1ms and - permitting 
lots of sodium ions (Na+) inside the cell - drive the membrane potential 
further until ~+20mV - this is the action potential or "spike" (not to 
be confused whith what we sometimes call spikes in EEG). The action 
potential is self-limiting bc. firstly, Na+channels also inactivate very 
quickly, and secondly, voltage-gated potassium channels are also opened 
by the depolarisation (they are just a bit slower), and permit potassium 
ions (K+) to flow out of the cell, thereby driving the membrane 
potential down again (repolarization, incl. "overshoot" below ~-70mV 
called hyperpolarization). Those action potentials are initiated 
(mostly) only in or close to the soma - "close" meaning a part of the 
axon, that is still very close to the soma and called axon-initial 
segment. The action potential is then conducted along all of the axon, 
which is the output compartment of the neuron, until it reaches its 
terminals, the presynapses, which then depolarize and thereby open 
voltage-dependent calcium channels, that trigger release of 
neurotransmitter onto the postsynapse of the next neuron in the chain.

Hope this answered your question - let me know, if not. The math of all 
this, is probably best explained in Christof Koch's book "The Biophysics 
of computation" - but I don't know, if you want to invest time & money 
into this, as this is the single neuron level and thus far away from the 
level we are normally dealing with.

best wishes,
Dennis

Kevan Hashemi wrote:
> Dear Dennis,
>
> If a pyramidal neuron is to generate a current circulation, its 
> post-synaptic membrane must be far from its soma. I believe the soma 
> is where the excitory current emerges from the cell, and the 
> post-synaptic membrane is where it goes in.
>
> If that's the case, then the dendrite holding the synaps must be at 
> the end of the long axon emerging from the cell. Does this mean that 
> axons can hold synapses that receive excitory currents? If so, does 
> this mean that axons carry the action potential out from the soma as 
> well as carrying excitory currents into the soma?
>
> Which bit of the neuron is the axon in any one of these pictures?
>
> http://www.nature.com/nrn/journal/v9/n3/fig_tab/nrn2286_F1.html
>
> Where is the synaps that receives a signal from another neuron?
>
> Thank you for answering my questions.
>
> Yours, Kevan
>

-- 
Dennis Kaetzel, D.Phil
Dept. of Physiology, Anatomy and Genetics
University of Oxford
Sherrington Building
Parks Road
Oxford
OX1 3PT