Average neurotransmitter count in a cell?

[icakeov]

Is there a general average number of neurotransmitters a given cell would "work with"?

 

[DiracPool]

Is there a general average number of neurotransmitters a given cell would "work with"?

Most neurons, for input, are influenced by a variety of different neurotransmitter species (namely, glutamate, GABA, the monoamines, and others) but project only a single neurotransmitter via their axon as output.

 

[icakeov]

Thanks @DiracPool ! I actually didn't know that, really good piece of information!
I was specifically wondering about actual numbers of neurotransmitters that it generally takes to create action potentials (rather than types of neurotransmitters) and whether the numbers are in the order of hundreds, or thousands or much more than that. Generally speaking of course.
Thanks again!

 

[DiracPool]

It's many more than hundreds or thousands or even hundreds of thousands. Also, it's not about the actual quantity or number of neurotransmitter molecules per se that influence whether an action potential manifests or not, it's the ratio of excitatory to inhibitory neurotransmitters that is important, plus other metabolic states of the neuron not related to neurotransmitters per se.

 

[jim mcnamara]

Hmm. Glutamate is known to cause spill-over synaptic crosstalk. That what you mean by 'more than hundreds or thousands ...'?

And FWIW motor neurons do not have massive crosstalk, AFAIK. So they may not fit what you describe, exactly. @DiracPool Correct me where I am wrong, please.

 

[DiracPool]

Hmm. Glutamate is known to cause spill-over synaptic crosstalk. That what you mean by 'more than hundreds or thousands ...'?

What I mean is that, if we look at a typical pyramidal neuron is the cerebral cortex, we see that, on average, it receives about 10,000 pre-synaptic inputs on it's apical dendrites from input neurons coming mainly from the thalamus or other regions of the cortex. The typical method of signaling between neurons through the synaptic cleft is for the pre-synaptic bouton to release synaptic vesicles. Each vesicle contains about 10,000 neurotransmitter molecules so, if you do the math, it's well into the millions or even billions of actual molecules that participate in any given action potential. However, again, the number can vary dramatically depending on the balance of inhibitory and excitatory species competing at any given time instant.

https://www.ncbi.nlm.nih.gov/books/NBK10866/

 

[icakeov]

That is so interesting!
The only thing I am confused about is the "pre-synaptic inputs" being on dendrites. Aren't dendrites supposed to be post-synaptic?

And also, does the full cycle of endocytosis, neurotransmitter uptake and then release (and anything else that I missed) really happen super fast and up to two times per second? (http://aiimpacts.org/rate-of-neuron-firing/)
If yes, that is extraordinary.

And just to double-check, would a "spike-train" then mean the progression of this action potential process as it moves from one area of the neuron to the other?

Many thanks again!

 

[BillTre]

Neuron diversity is great, especially if you include the neurons of invertebrates. As a result there is no simple answer to how many neurotransmitter molecules it takes to trigger a neuron to make an action potential.

The relationship between transmitter amounts affecting a particular neuron and its production of action potentials depends a lot on the geometry and physiological properties of the neuron, as well as where its inputs are located on the neuron.
Also not all neurons even generate action potentials. Some are non-spiking, just having graded changes in membrane potential.

Among neurons producing action potentials, there is usually a single critical point (the axon hillock) where the graded changes in membrane potential resulting from synaptic inputs will either exceed a threshold and trigger an action potential or fail a dissipate.

How the various inputs, as discussed above, sum up together is often analyzed by modeling with the neuron's cable properties. This takes into consideration many properties of the neuron and where its various inputs are. Among the neuron's properties considered are the length and electrical properties (capacitance of membrane, how well different parts are electrically connected), of the number, size and length of neuronal branches (such as dendrites) as well as the cell body, location, density and type of synaptic inputs. This allows the neuron to be modeled in a geometrically simpler form in an attempt to figure out what it takes to generate action potentials in particular neurons.

https://www.amazon.com/dp/019515956X/?tag=pfamazon01-20 is a classic book on this approach that explains this stuff better than I could. Here is a pdf of the first chapter which explains a several concepts.

 

[icakeov]

Many thanks BillTre! So many great references to go to.